Communication method, communication apparatus, and communication system

Abstract

Provided in the present application are a communication method, a communication apparatus, and a communication system. The method provides a device-level QP establishment process, in which a first device and a second device pre-establish a device-level QP and an RDMA link on the basis of an indication of a control-plane network element, or a first device and a second device negotiate with each other on the basis of a QoS requirement provided by the control-plane network element to establish the QP and the RDMA link. In the case that the RDMA link has been established in advance, in a session establishment process, the first device and the second device then use a certain RDMA link, which matches the QoS requirement of a service, among pre-established RDMA links to perform service transmission, such that the first device and the second device can allocate data packets of services having different QoS requirements to corresponding QPs for transmission, thereby realizing QoS mapping when RDMA transmission is used in a 5G network or a future communication network, and further realizing a QoS guarantee.

Description

Communication method, communication device and communication system

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to the Chinese patent application filed with the State Intellectual Property Office of the People's Republic of China on December 25, 2023, with application number 202311797597.2 and invention name "Communication Method, Communication Device and Communication System", the entire contents of which are incorporated by reference into this application. Technical Field

[0003] The present application relates to the field of wireless communication technology, and in particular to a communication method, a communication device, and a communication system. Background Art

[0004] In the current fifth-generation (5G) user-plane data transmission, data transmission between different nodes requires significant participation of the central processing unit (CPU) in the data transmission process. However, this data transmission solution often encounters computer system bottlenecks caused by the mismatch between CPU and memory speeds, often referred to as the "memory wall." This is because CPU speeds are increasing, while memory speeds have not kept pace. Data transmission between the CPU and memory takes time. When the CPU needs to read or write large amounts of data, it waits for memory, resulting in the CPU being unable to fully utilize its computing power, affecting the overall performance of the computer system. This data transmission method, which requires significant CPU participation, is also known as messaging through the kernel. Because this data transmission method requires kernel transmission, it incurs high data movement and copying overhead. Furthermore, this data transmission method may not meet the high throughput and low latency requirements of certain services.

[0005] To reduce the CPU overhead of data transmission in 5G networks and meet the high-throughput and low-latency requirements of businesses, remote direct memory access (RDMA) transmission can be introduced into 5G networks. This involves establishing an RDMA link between two nodes in the 5G network for data transmission. RDMA technology offers the following advantages: 1) Zero copy, meaning that data does not need to be copied back and forth between user space and kernel space. 2) Kernel bypass, meaning that I / O data flows can bypass the kernel. Data can be prepared and the hardware notified at the user level to prepare for transmission and reception, avoiding the overhead of system calls and context switches. 3) CPU offloading, meaning that memory can be read and written without the remote node's CPU participating in the communication (of course, this requires the "key" to access the remote memory segment). Therefore, RDMA technology can achieve high throughput, low latency, and reduced CPU load for network transmission.

[0006] However, the quality of service (QoS) mapping mechanism in 5G networks or future communication networks is somewhat different from the QoS mechanism in RDMA. Therefore, when using RDMA transmission in 5G networks or future communication networks, how to implement RDMA QoS mapping remains to be solved.

[0007] Summary of the Invention

[0008] The present application provides a communication method, a communication device, and a communication system for implementing RDMA QoS mapping when using RDMA transmission in a 5G network or a future communication network, thereby achieving QoS guarantee.

[0009] In a first aspect, an embodiment of the present application provides a communication method, which can be executed by a first device or a module (such as a chip) of the first device, wherein the first device is a user plane network element or an access network device. The method includes: receiving indication information from a control plane network element, the indication information including queue pair (QP) information of the first device, the indication information being used to indicate establishment of a QP based on the QP information of the first device, the QP information of the first device including at least one of the following information: the number of QPs to be established, the memory area corresponding to the QP to be established, the access rights of the QP to be established, the type of the QP to be established, the maximum transmission unit of the QP to be established, the message length of the transmission of the QP to be established, or the mapping relationship between the QP to be established and the quality of service (QoS) requirement; and establishing at least one QP according to the indication information.

[0010] The above solution proposes a device-level QP establishment process, in which the first device pre-establishes a device-level QP based on the instructions of the control plane network element, and can also establish an RDMA link based on the pre-established QP. When the RDMA link is pre-established, during the session establishment process, the first device uses one of the pre-established RDMA links that is adapted to the QoS requirements of the service to transmit the service. This allows the first device to allocate data packets of services with different QoS requirements to corresponding QPs for transmission, realizing QoS mapping when using RDMA transmission in 5G networks or future communication networks, thereby achieving QoS assurance.

[0011] In a possible implementation method, the indication information is further used to instruct establishment of an RDMA link between the first device and the second device; and the RDMA link between the first device and the second device is established according to the at least one QP.

[0012] In one possible implementation method, the QoS requirement includes at least one of the following information: RDMA connection type, RDMA operation type, RDMA transmission type, RDMA protocol type, service latency requirement, service bandwidth requirement, service priority requirement or transmission rate.

[0013] In a possible implementation method, the method further includes: sending QP establishment information of the first device to the control plane network element, where the QP establishment information of the first device is used to indicate information of the QP actually established by the first device.

[0014] In one possible implementation method, the method further includes: receiving a packet detection rule (PDR) from the control plane network element, the PDR including service feature information of the first service, the service feature information including at least one of the following information: RDMA connection type, RDMA operation type, RDMA transmission type, RDMA protocol type, service latency requirement, service bandwidth requirement, service priority requirement, or transmission rate; determining a first QoS requirement in the mapping relationship that matches the service feature information; determining, based on the mapping relationship, a first QP among the at least one QP that corresponds to the first QoS requirement; and using the first QP to transmit a data packet that matches the service feature information.

[0015] In the above solution, when the RDMA link is established in advance, in the session establishment process, the control plane network element sends a PDR to the first device. The first device selects a suitable QP for service transmission based on the PDR and the mapping relationship between QP and QoS requirements. In this way, the first device can allocate data packets of services with different QoS requirements to the corresponding QP for transmission, thereby realizing QoS mapping when using RDMA transmission in 5G networks or future communication networks, and further realizing QoS guarantee.

[0016] In one possible implementation method, the method further includes: receiving a notification message from the control plane network element, the notification message including information of a first QP among the at least one QP, the notification message being used to indicate the use of the first QP to transmit data packets of the first service; and according to the notification message, using the first QP to transmit data packets of the first service.

[0017] In the above solution, when the RDMA link is established in advance, in the session establishment process, the control plane network element selects a suitable QP for the first device for service transmission based on the QoS requirements and the mapping relationship between QP and QoS requirements. In this way, the first device can allocate data packets of services with different QoS requirements to the corresponding QP for transmission, thereby realizing QoS mapping when using RDMA transmission in 5G networks or future communication networks, and further realizing QoS guarantee.

[0018] In a possible implementation method, the method further includes: sending feedback information to the control plane network element, where the feedback information is used to indicate a transmission status of the RDMA link corresponding to the first QP.

[0019] In the above solution, the first device feeds back the transmission status of the RDMA link to the control plane network element, so that the control plane network element can grasp the transmission status of the RDMA link in real time and update the QP accordingly, thereby actively improving the link transmission quality.

[0020] In a second aspect, an embodiment of the present application provides a communication method, which can be executed by a control plane network element or a module (such as a chip) of the control plane network element, where the control plane network element is a session management network element or a policy control network element. The method includes: sending first indication information to a first device, the first indication information including QP information of the first device, the first indication information being used to instruct establishment of a QP based on the QP information of the first device, the QP information of the first device including at least one of the following information: the number of QPs to be established, a memory area corresponding to the QP to be established, access rights of the QP to be established, a type of the QP to be established, a maximum transmission unit of the QP to be established, a message length of transmission of the QP to be established, or a mapping relationship between the QP to be established and a QoS requirement; and sending second indication information to a second device, the second indication information including the QP information of the second device, the second indication information being used to instruct establishment of a QP based on the QP information of the second device, the QP information of the second device including at least one of the following information: the number of QPs to be established, a memory area corresponding to the QP to be established, access rights of the QP to be established, a type of the QP to be established, a maximum transmission unit of the QP to be established, a message length of transmission of the QP to be established, or a mapping relationship between the QP to be established and a QoS requirement.

[0021] The above solution proposes a device-level QP establishment process, in which the first device and the second device pre-establish a device-level QP based on the instructions of the control plane network element, and can also establish an RDMA link based on the pre-established QP. When the RDMA link is pre-established, during the session establishment process, the first device and the second device use one of the pre-established RDMA links that is adapted to the QoS requirements of the service to transmit the service. In this way, the first device and the second device can allocate data packets of services with different QoS requirements to corresponding QPs for transmission, realizing QoS mapping when using RDMA transmission in 5G networks or future communication networks, and thus achieving QoS guarantee.

[0022] In a possible implementation method, the method further includes: determining the QP information of the first device based on the information of the first device; wherein the information of the first device includes at least one of the hardware information of the first device, the memory information of the first device, or the RDMA information of the first device, and the RDMA information of the first device includes the RDMA protocol supported by the first device and the bearer network type corresponding to the supported RDMA protocol.

[0023] In a possible implementation method, the method further includes: determining the QP information of the second device based on the information of the second device; wherein the information of the second device includes at least one of the hardware information of the second device, the memory information of the second device, or the RDMA information of the second device, and the RDMA information of the second device includes the RDMA protocol supported by the second device and the bearer network type corresponding to the supported RDMA protocol.

[0024] In a possible implementation method, the method further includes: receiving QP establishment information of the first device from the first device, where the QP establishment information of the first device is used to indicate information of the QP actually established by the first device.

[0025] In a possible implementation method, the method further includes: updating the QP information of the first device according to the QP establishment information of the first device.

[0026] In a possible implementation method, the method further includes: receiving QP establishment information of the second device from the second device, where the QP establishment information of the second device is used to indicate information of the QP actually established by the second device.

[0027] In a possible implementation method, the method further includes: updating the QP information of the second device according to the QP establishment information of the second device.

[0028] In one possible implementation method, the method further includes: receiving feedback information from the first device or the second device, where the feedback information is used to indicate a transmission status of an RDMA link corresponding to a QP used when the first device and the second device perform data transmission; and sending an update indication to the first device or the second device based on the feedback information, where the update indication is used to instruct an update of the used QP.

[0029] In the above solution, the first device or the second device feeds back the transmission status of the RDMA link to the control plane network element, so that the control plane network element can grasp the transmission status of the RDMA link in real time and update the QP accordingly, thereby effectively improving the link transmission quality.

[0030] In a third aspect, an embodiment of the present application provides a communication method, which can be performed by a first device or a module (such as a chip) of the first device, where the first device is a user plane network element or an access network device. The method includes: receiving indication information from a control plane network element, the indication information including at least one QoS requirement, the indication information being used to instruct establishment of a QoS requirement that satisfies the at least one QoS requirement; establishing, based on the indication information, at least one QoS requirement that satisfies the at least one QoS requirement, and establishing a mapping relationship between the at least one QoS requirement and the at least one QP.

[0031] The above solution proposes a device-level QP establishment process. Based on the QoS requirements provided by the control plane network element, the first device negotiates with the second device to establish a QP and, in addition, an RDMA link. After the RDMA link is pre-established, during the session establishment process, the first and second devices use one of the pre-established RDMA links that is adapted to the QoS requirements of the service to transmit the service. This allows the first device to allocate data packets for services with different QoS requirements to the corresponding QP for transmission, thus achieving QoS mapping when using RDMA transmission in 5G networks or future communication networks, thereby achieving QoS assurance.

[0032] In a possible implementation method, the indication information is further used to instruct establishment of an RDMA link between the first device and the second device; the method further includes: establishing the RDMA link between the first device and the second device according to the at least one QP.

[0033] In one possible implementation method, the QoS requirement includes at least one of the following information: RDMA connection type, RDMA operation type, RDMA transmission type, RDMA protocol type, service latency requirement, service bandwidth requirement, service priority requirement or transmission rate.

[0034] In one possible implementation method, the method further includes: sending QP establishment information of the first device to the control plane network element, where the QP establishment information of the first device is used to indicate information about the QP actually established by the first device, and the information about the QP actually established by the first device includes the mapping relationship.

[0035] In a possible implementation method, the method further includes: receiving a PDR from the control plane network element, the PDR including service feature information of the first service, the service feature information including at least one of the following information: RDMA connection type, RDMA operation type, RDMA transmission type, RDMA protocol type, service latency requirement, service bandwidth requirement, service priority requirement, or transmission rate; determining a first QoS requirement in the mapping relationship that matches the service feature information; determining, based on the mapping relationship, a first QP corresponding to the first QoS requirement among the at least one QP; and using the first QP to transmit a data packet that matches the service feature information.

[0036] In the above solution, when the RDMA link is established in advance, in the session establishment process, the control plane network element sends a PDR to the first device. The first device selects a suitable QP for service transmission based on the PDR and the mapping relationship between QP and QoS requirements. In this way, the first device can allocate data packets of services with different QoS requirements to the corresponding QP for transmission, thereby realizing QoS mapping when using RDMA transmission in 5G networks or future communication networks, and further realizing QoS guarantee.

[0037] In one possible implementation method, the method further includes: receiving a notification message from the control plane network element, the notification message including information of a first QP among the at least one QP, the notification message being used to indicate the use of the first QP to transmit data packets of the first service; and according to the notification message, using the first QP to transmit data packets of the first service.

[0038] In the above solution, when the RDMA link is established in advance, in the session establishment process, the control plane network element selects a suitable QP for the first device for service transmission based on the QoS requirements and the mapping relationship between QP and QoS requirements. In this way, the first device can allocate data packets of services with different QoS requirements to the corresponding QP for transmission, thereby realizing QoS mapping when using RDMA transmission in 5G networks or future communication networks, and further realizing QoS guarantee.

[0039] In a possible implementation method, the method further includes: sending feedback information to the control plane network element, where the feedback information is used to indicate a transmission status of the RDMA link corresponding to the first QP.

[0040] In the above solution, the first device feeds back the transmission status of the RDMA link to the control plane network element, so that the control plane network element can grasp the transmission status of the RDMA link in real time and update the QP accordingly, thereby actively improving the link transmission quality.

[0041] In a fourth aspect, an embodiment of the present application provides a communication method, which can be performed by a control plane network element or a module (such as a chip) of the control plane network element, wherein the control plane network element is a session management network element or a policy control network element. The method includes: sending first indication information to a first device, the first indication information including at least one QoS requirement, the first indication information being used to instruct the establishment of a QP that meets the at least one QoS requirement; and sending second indication information to a second device, the second indication information including the at least one QoS requirement, the second indication information being used to instruct the establishment of a QP that meets the at least one QoS requirement.

[0042] The above solution proposes a device-level QP establishment process, where the first and second devices independently negotiate to establish a QP and establish an RDMA link based on the QoS requirements provided by the control plane network element. After the RDMA link is pre-established, during the session establishment process, the first and second devices use one of the pre-established RDMA links that is adapted to the QoS requirements of the service to transmit the service. This allows the first and second devices to allocate data packets for services with different QoS requirements to corresponding QPs for transmission, thus achieving QoS mapping when using RDMA transmission in 5G networks or future communication networks, thereby achieving QoS assurance.

[0043] In one possible implementation method, sending the first indication information to the first device includes: sending the first indication information to the first device based on the information of the first device; wherein the information of the first device includes at least one of the hardware information of the first device, the memory information of the first device, or the RDMA information of the first device, and the RDMA information of the first device includes the RDMA protocol supported by the first device and the bearer network type corresponding to the supported RDMA protocol.

[0044] In one possible implementation method, sending the second indication information to the second device includes: sending the second indication information to the second device based on the information of the second device; wherein the information of the second device includes at least one of the hardware information of the second device, the memory information of the second device, or the RDMA information of the second device, and the RDMA information of the second device includes the RDMA protocol supported by the second device and the bearer network type corresponding to the supported RDMA protocol.

[0045] In one possible implementation method, the method further includes: receiving QP establishment information of the first device from the first device, the QP establishment information of the first device is used to indicate information of the QP actually established by the first device, and the information of the QP actually established by the first device includes a mapping relationship between the at least one QoS requirement and the at least one QP established by the first device.

[0046] In a possible implementation method, the method further includes: updating the QP information of the first device according to the QP establishment information of the first device.

[0047] In one possible implementation method, the method further includes: receiving QP establishment information of the second device from the second device, the QP establishment information of the second device is used to indicate information of the QP actually established by the second device, and the information of the QP actually established by the second device includes a mapping relationship between the at least one QoS requirement and the at least one QP established by the second device.

[0048] In a possible implementation method, the method further includes: updating the QP information of the second device according to the QP establishment information of the second device.

[0049] In one possible implementation method, the method further includes: receiving feedback information from the first device or the second device, where the feedback information is used to indicate a transmission status of an RDMA link corresponding to a QP used when the first device and the second device perform data transmission; and sending an update indication to the first device or the second device based on the feedback information, where the update indication is used to instruct an update of the used QP.

[0050] In the above solution, the first device or the second device feeds back the transmission status of the RDMA link to the control plane network element, so that the control plane network element can grasp the transmission status of the RDMA link in real time and update the QP accordingly, thereby effectively improving the link transmission quality.

[0051] In a fifth aspect, an embodiment of the present application provides a communication device, which may be a first device or a module (such as a chip) of the first device. The device has the function of implementing any implementation method of the first or third aspects described above. The function can be implemented by hardware or by hardware executing corresponding software implementations. The hardware or software includes one or more modules corresponding to the above functions.

[0052] In a sixth aspect, an embodiment of the present application provides a communication device, which may be a control plane network element or a module (such as a chip) of a control plane network element. The device has the function of implementing any implementation method of the second or fourth aspects above. The function can be implemented by hardware or by hardware executing corresponding software. The hardware or software includes one or more modules corresponding to the above functions.

[0053] In a seventh aspect, an embodiment of the present application provides a communication device, comprising a unit or means for executing each step of any implementation method in the above-mentioned first to fourth aspects.

[0054] In an eighth aspect, an embodiment of the present application provides a communication device, comprising a processor and an interface circuit, wherein the processor is configured to communicate with other devices via the interface circuit and execute any of the implementation methods described in the first to fourth aspects. The processor comprises one or more.

[0055] In a ninth aspect, an embodiment of the present application provides a communication device, comprising a processor, the processor being configured to call a program to execute any of the implementation methods in the first to fourth aspects above. The processor may be one or more.

[0056] Optionally, the communication device may further include a memory, which is coupled to the processor and may be located inside or outside the device.

[0057] In the tenth aspect, an embodiment of the present application provides a communication device, comprising a processor; when the device is running, the processor executes computer instructions to enable the device to execute any implementation method in the above-mentioned first to fourth aspects.

[0058] Optionally, the communication device may further include a memory for storing the computer instructions.

[0059] In the eleventh aspect, an embodiment of the present application further provides a computer program product, which includes a computer program or instructions. When the computer program or instructions are run by a communication device, any implementation method in the above-mentioned first to fourth aspects is executed.

[0060] In the twelfth aspect, an embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, which, when run on a communication device, enables any implementation method in the above-mentioned first to fourth aspects to be executed.

[0061] In the thirteenth aspect, an embodiment of the present application further provides a chip system, comprising: a processor for executing any implementation method in the above-mentioned first to fourth aspects.

[0062] In the fourteenth aspect, an embodiment of the present application also provides a communication system, including a control plane network element and a first device; the control plane network element is used to send a first indication message to the first device, the first indication message includes the QP information of the first device, and the first indication message is used to indicate the establishment of a QP based on the QP information of the first device, and the QP information of the first device includes at least one of the following information: the number of queue pairs QP to be established, the memory area corresponding to the QP to be established, the access rights of the QP to be established, the type of the QP to be established, the maximum transmission unit of the QP to be established, the transmission message length of the QP to be established, or the mapping relationship between the QP to be established and the QoS requirement; the first device is used to receive the first indication message; and establish at least one QP according to the first indication information.

[0063] In one possible implementation method, the communication system also includes a second device; the control plane network element is further used to send second indication information to the second device, the second indication information includes QP information of the second device, the second indication information is used to indicate the establishment of a QP based on the QP information of the second device, and the QP information of the second device includes at least one of the following information: the number of queue pairs QP to be established, the memory area corresponding to the QP to be established, the access rights of the QP to be established, the type of the QP to be established, the maximum transmission unit of the QP to be established, the message length of the transmission of the QP to be established, or the mapping relationship between the QP to be established and the QoS requirement; the second device is used to receive the second indication information; and establish at least one QP according to the second indication information.

[0064] In one possible implementation method, the first indication information is further used to indicate establishment of an RDMA link between the first device and the second device, and the second indication information is further used to indicate establishment of an RDMA link between the second device and the first device; the first device is further used to establish at least one RDMA link with the second device based on at least one QP established by the first device; and the second device is further used to establish at least one RDMA link with the first device based on at least one QP established by the second device.

[0065] In the fifteenth aspect, an embodiment of the present application also provides a communication system, including a control plane network element and a first device; the control plane network element is used to send a first indication message to the first device, the first indication message including at least one quality of service QoS requirement, and the first indication message is used to indicate the establishment of a queue pair QP that meets the at least one QoS requirement; the first device is used to receive the first indication message; according to the first indication message, establish at least one QP that meets the at least one QoS requirement, and establish a mapping relationship between the at least one QoS requirement and the at least one QP.

[0066] In one possible implementation method, the communication system also includes a second device; the control plane network element is further used to send second indication information to the second device, the second indication information includes the at least one QoS requirement, and the second indication information is used to indicate the establishment of a QP that meets the at least one QoS requirement; the second device is used to receive the second indication information; according to the second indication information, establish at least one QP that meets the at least one QoS requirement, and establish a mapping relationship between the at least one QoS requirement and the at least one QP.

[0067] In one possible implementation method, the first indication information is further used to indicate establishment of an RDMA link between the first device and the second device, and the second indication information is further used to indicate establishment of an RDMA link between the second device and the first device; the first device is further used to establish at least one RDMA link with the second device based on at least one QP established by the first device; and the second device is further used to establish at least one RDMA link with the first device based on at least one QP established by the second device. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 is a schematic diagram of a 5G network architecture based on a service-oriented architecture;

[0069] Figure 2 is a schematic diagram of memory access without DMA;

[0070] Figure 3 is a schematic diagram of memory access with DMA;

[0071] Figure 4 is a schematic diagram of memory access between different nodes in a traditional network;

[0072] Figure 5 is a schematic diagram of memory access between different nodes when RDMA is enabled;

[0073] FIG6 is an example diagram of QoS mapping in 5G;

[0074] FIG7 is an example diagram of a virtual channel in RDMA;

[0075] FIG8( a ) is a flow chart of a communication method according to an embodiment of the present application;

[0076] FIG8( b ) is a flow chart of a communication method according to an embodiment of the present application;

[0077] 9 to 12 are flowcharts of a communication method according to an embodiment of the present application;

[0078] 13 and 14 are schematic diagrams of communication devices provided in embodiments of the present application. DETAILED DESCRIPTION

[0079] To meet the challenges of wireless broadband technology and maintain the leading edge of the 3rd Generation Partnership Project (3GPP) network, the 3GPP standards group has developed a next-generation mobile communications network system architecture, known as the fifth-generation (5G) network architecture. This architecture not only supports access to the 5G core network (CN) using 3GPP-defined radio access technologies (such as Long Term Evolution (LTE) and 5G Radio Access Network (RAN)), but also supports access to the core network using non-3GPP access technologies via the non-3GPP interworking function (N3IWF) or the next-generation packet data gateway (ngPDG).

[0080] Figure 1 is a schematic diagram of a 5G network architecture based on a service-oriented architecture. The 5G network architecture shown in Figure 1 may include access network equipment and core network equipment. Terminal equipment accesses the data network (DN) through access network equipment and core network equipment. The core network equipment includes but is not limited to some or all of the following network elements: authentication server function (AUSF) network element, unified data management (UDM) network element, unified data repository (UDR) network element, network repository function (NRF) network element, network exposure function (NEF) network element, application function (AF) network element, policy control function (PCF) network element, access and mobility management function (AMF) network element, session management function (SMF) network element, and user plane function (UPF) network element.

[0081] The terminal device can be user equipment (UE), a mobile station, a mobile terminal device, etc. The terminal device can be widely used in various scenarios, for example, device-to-device (D2D), vehicle to everything (V2X) communication, machine-type communication (MTC), Internet of Things (IOT), virtual reality, augmented reality, industrial control, autonomous driving, telemedicine, smart grid, smart furniture, smart office, smart wearable, smart transportation, smart city, etc. The terminal device can be a mobile phone, a tablet computer, a computer with wireless transceiver function, a wearable device, a vehicle, an urban air vehicle (such as an unmanned aerial vehicle, a helicopter, etc.), a ship, a robot, a robotic arm, a smart home device, etc. For the sake of convenience, this application uses UE as an example of a terminal device for illustration, and any UE appearing in any subsequent position can be replaced by a terminal device.

[0082] Access network equipment can be wireless access network equipment or wired access network equipment. Wireless access network equipment includes 3GPP access network equipment, untrusted non-3GPP access network equipment, and trusted non-3GPP access network equipment. 3GPP access network equipment includes, but is not limited to, evolved NodeBs (eNodeBs) in LTE, next-generation NodeBs (gNBs) in 5G mobile communication systems, base stations in future mobile communication systems, or modules or units that perform some of the functions of base stations, such as centralized units (CUs) and distributed units (DUs). Untrusted non-3GPP access network equipment includes, but is not limited to, untrusted non-3GPP access gateways or N3IWFs, untrusted wireless local area network (WLAN) access points (APs), switches, and routers. Trusted non-3GPP access network equipment includes, but is not limited to, trusted non-3GPP access gateways, trusted WLAN APs, switches, and routers. Wired access network equipment includes, but is not limited to, wireline access gateways, fixed-line network equipment, switches, and routers. For ease of explanation, this application uses a base station as an example of an access network device, and any base station appearing at any subsequent location can be replaced by an access network device.

[0083] Base stations and UEs can be fixed or mobile. They can be deployed on land, indoors or outdoors, handheld or vehicle-mounted; on water; or in the air on aircraft, balloons, and satellites. The embodiments of this application do not limit the application scenarios of base stations and UEs.

[0084] The AMF network element performs functions such as mobility management and access authentication / authorization. It is also responsible for transferring user policies between the UE and the PCF.

[0085] The SMF network element includes functions such as performing session management, executing control policies issued by the PCF network element, selecting the UPF network element, or allocating the UE's Internet Protocol (IP) address.

[0086] The UPF network element includes functions such as user plane data forwarding, session / flow-level billing statistics, or bandwidth limitation.

[0087] UDM network elements include functions such as executing and managing contract data or user access authorization.

[0088] UDR includes functions for accessing data such as contract data, policy data, or application data.

[0089] NEF network element is used to support the opening of capabilities and events.

[0090] The AF network element communicates application-side requirements to the network, such as Quality of Service (QoS) requirements or user status event subscriptions. The AF can be a third-party functional entity or an application service deployed by an operator, such as the IP Multimedia Subsystem (IMS) voice call service. AF network elements include those within the core network (i.e., the operator's AF network element) and third-party AF network elements (such as an enterprise's application server).

[0091] The PCF network element includes policy control functions such as billing for sessions and service flow levels, QoS bandwidth guarantee and mobility management, or UE policy decision-making. PCF network elements include access and mobility management policy control function (AM PCF) network element and session management policy control function (SM PCF) network element. Among them, the AM PCF network element is used to formulate AM policy and user policy for UE. The AM PCF network element can also be called a policy control network element that provides services for UE (PCF for a UE). The SM PCF network element is used to formulate session management policy (SMpolicy) for the session. The SM PCF network element can also be called a policy control network element that provides services for protocol data unit (PDU) sessions ((PCF for a PDU session))).

[0092] NRF network elements can be used to provide network element discovery functions, providing network element information corresponding to the network element type based on requests from other network elements. NRF network elements also provide network element management services, such as network element registration, update, deregistration, or network element status subscription and push.

[0093] The AUSF network element is responsible for authenticating users to determine whether users or devices are allowed to access the network.

[0094] A DN is a network located outside of a carrier network. A carrier network can connect to multiple DNs, and a variety of services can be deployed on the DN, providing UEs with data and / or voice services. For example, a DN is the private network of a smart factory. Sensors installed in the workshop can be UEs. The DN houses a control server for these sensors, which can provide services to the sensors. The sensors can communicate with the control server, receive instructions from the control server, and transmit collected sensor data to the control server based on the instructions. Another example is a DN that is a company's internal office network. An employee's mobile phone or computer can be a UE, allowing them to access information and data resources on the company's internal office network.

[0095] In Figure 1, Nausf, Npcf, Nudr, Nudm, Naf, Namf, Nsmf, Nnef, and Nnrf are the service-based interfaces (SBIs) provided by the aforementioned AUSF, PCF, UDR, UDM, AF, AMF, SMF, NEF, and NRF, respectively, and are used to invoke corresponding service-based operations. N1, N2, N3, N4, and N6 are interface serial numbers, and their meanings are as follows:

[0096] 1) N1: The interface between the AMF network element and the UE, which can be used to deliver non-access stratum (NAS) signaling (such as QoS rules from the AMF network element) to the UE.

[0097] 2) N2: The interface between the AMF network element and the base station, which can be used to transmit radio bearer control information from the core network side to the base station.

[0098] 3) N3: The interface between the base station and the UPF network element, mainly used to transmit uplink and downlink user plane data between the base station and the UPF network element.

[0099] 4) N4: The interface between the SMF network element and the UPF network element, which can be used to transmit information between the control plane and the user plane, including controlling the issuance of forwarding rules, QoS rules, traffic statistics rules, etc. for the user plane and reporting information on the user plane.

[0100] 5) N6: The interface between UPF network element and DN, used to transmit uplink and downlink user data flows between UPF network element and DN.

[0101] The various network function elements in the architecture shown in Figure 1 are connected through a service-based bus and interact through service-based interfaces. The advantages of the service-based bus are that it improves the flexibility, openness, scalability and intelligence of the network, and can support diverse business scenarios and needs. The service-based bus can be used to transmit various types of data and signaling. For example, it can be used to transmit real-time signaling that is sensitive to latency (such as service-based interface call signaling between network element function elements), it can also be used to transmit real-time data that is sensitive to latency (such as real-time artificial intelligence inference data), and it can also be used to transmit non-real-time data (such as offline artificial intelligence training data). Moreover, when the service-based bus transmits these data or signaling, these data or signaling are coupled together, that is, the service-based bus can be used for the transmission of real-time signaling, real-time data and non-real-time data at the same time.

[0102] It is understood that the above-mentioned network element or function can be a network element in a hardware device, a software function running on dedicated hardware, or a virtualized function instantiated on a platform (e.g., a cloud platform). Optionally, the above-mentioned network element or function can be implemented by a single device, or by multiple devices, or can be a functional module within a single device, and this is not specifically limited in the embodiments of the present application.

[0103] The user plane network element, policy control network element, and session management network element in this application can be the UPF network element, PCF network element, and SMF network element in Figure 1, respectively, or can be a network element having the functions of the above-mentioned UPF network element, PCF network element, and SMF network element in future communications such as 6G networks. This application is not limited to this. In the embodiments of this application, an example is used to describe the UPF network element, PCF network element, and SMF network element as the user plane network element, policy control network element, and session management network element, respectively, and the UPF network element, PCF network element, and SMF network element are abbreviated as UPF, PCF, and SMF, respectively.

[0104] To facilitate understanding of the content of this application, the background technology involved in this application is first introduced below.

[0105] 1. Direct memory access (DMA)

[0106] DMA refers to the process of external devices reading and writing memory directly without the participation of the central processing unit (CPU).

[0107] Figure 2 illustrates memory access without DMA. Assuming the input / output (I / O) device is a standard network interface card (NIC), to retrieve the data to be sent from memory and then assemble the data packet to send to the physical link, the NIC needs to notify the CPU of its data request via the bus. Based on this data request, the CPU then copies the corresponding data from the memory buffer to its internal registers and then to the I / O device's storage space. If the data volume is large, the CPU will be busy moving data for a long time, unable to devote time to other tasks. The CPU's primary task is computation, not data copying. Therefore, this data copying wastes the CPU's computing power. To reduce the CPU's workload and allow it to focus on more meaningful tasks, the DMA mechanism was designed.

[0108] Figure 3 shows a schematic diagram of memory access with DMA. As can be seen, a DMA controller is attached to the bus, specifically for reading and writing memory. When the network card copies data from memory, the entire data copy process is completed by the DMA controller, except for some necessary control commands. This process is identical to the CPU copy shown in Figure 2, except that the DMA method copies the data from memory via the bus to registers within the DMA controller, and then to the storage space of the I / O device. The CPU, aside from monitoring the start and end of this process, can focus on other tasks. The DMA controller is typically located within the I / O device, meaning that the I / O device contains both a module responsible for data transmission and reception and a DMA module.

[0109] 2. Remote direct memory access (RDMA)

[0110] RDMA enables a local node to directly access the memory of a remote node. Direct access means that the local node can read and write to remote memory just like accessing local memory, bypassing the complex Transmission Control Protocol (TCP) / IP network stack of traditional Ethernet. This process is invisible to the other end, and most of the work in this read and write process is performed by hardware rather than software.

[0111] Figure 4 illustrates memory access between different nodes in a traditional network. In a traditional network, "node A sending a message to node B" actually means "moving a piece of data from node A's memory to node B's memory via a network link." This process, both sending and receiving, requires CPU command and control, including network interface card (NIC) control, interrupt processing, packet encapsulation, and parsing. The data in user space memory on the left node in Figure 4 must be copied by the CPU to a kernel buffer before it can be accessed by the NIC. During this process, the data passes through the software-implemented TCP / IP protocol stack, with headers and checksums added to various layers, such as the TCP header and IP header. The NIC copies the kernel data to its internal buffer via DMA, processes it, and then sends it to the peer via a physical link. Upon receiving the data, the peer performs the reverse process: copying the data from the NIC's internal storage space to a kernel buffer via DMA. The CPU then parses the data through the TCP / IP protocol stack, extracts the data, and copies it to user space. As can be seen, even with DMA technology, this process still relies heavily on the CPU.

[0112] Figure 5 illustrates memory access between different nodes when RDMA is enabled. Similarly, copying a segment of data from local memory to peer memory requires minimal involvement of the CPUs on either end (only the control plane) when using RDMA. The local RDMA network card directly copies data from user-space memory to the RDMA network card's internal storage space via DMA. The hardware then assembles the packets at each layer and sends them to the peer RDMA network card via the physical link. After receiving the data, the peer RDMA network card strips off the headers and checksums from each layer and copies the data directly to user-space memory via DMA.

[0113] RDMA technology has the following advantages:

[0114] 1) 0 copy means that there is no need to copy data back and forth between user space and kernel space.

[0115] 2) Kernel Bypass: This means that the I / O data flow can bypass the kernel, that is, the data can be prepared at the user level and the hardware can be notified to prepare for sending and receiving, avoiding the overhead of system calls and context switches.

[0116] 3) CPU offloading: This allows the remote node's CPU to read and write memory without participating in the communication (of course, this requires the "key" to access the remote memory segment). This effectively moves message encapsulation and parsing into the hardware. In traditional Ethernet communication, both CPUs must participate in message parsing at all layers. If the data volume is large and the exchange is frequent, this will result in a significant CPU overhead, and these CPU computing resources could be used for more valuable work.

[0117] Therefore, RDMA technology can achieve high throughput and low latency in network transmission and reduce CPU load.

[0118] The RDMA protocol stack has multiple versions, listed in chronological order: IB -> RoCEv1 -> iWARP -> RoCEv2. IB stands for Infiniband, RoCE stands for RDMA over converged Ethernet, and iWARP stands for Internet Wide Area RDMA protocol. RoCE includes version 1 (v1) and version 2 (v2).

[0119] RDMA uses work queues to queue a series of service requests for execution. Work queues are called queue pairs (QPs) in RDMA. One queue in the pair is for send operations, and the other queue is for receive operations. Generally speaking, the send work queue holds instructions that cause data to be transferred between the memory of a user and the memory of another user, while the receive work queue holds instructions on where to place data received from the other user. The other user is called the remote user, even though it may be located on the same node. RDMA supports both connection-oriented and datagram-oriented services. For connection-oriented services, each QP is associated with only one remote user. In this case, the QP context is configured to identify the queue pair of the remote user. The remote user is identified by a port and a queue pair number (QPN). The port is identified by a local ID (LID) or by a local ID and a global ID (GID).

[0120] RDMA includes but is not limited to the following four connection types:

[0121] 1) Reliable connection (RC): Provides message-oriented reliable transmission. The QPs at both ends of the communication are bound one-to-one. This is the most commonly used RDMA connection type.

[0122] 2) Unreliable connection (UC): provides message-oriented unreliable transmission, and the QPs at both ends of the communication are bound one by one.

[0123] 3) Unreliable Datagram (UD): This type of transport provides unreliable message-oriented transport, without a one-to-one binding between the two communicating peers. This transport type is very similar to the User Datagram Protocol (UDP).

[0124] 4) Reliable datagram (RD): Currently, most network card manufacturers do not support it.

[0125] 3. User Plane Data Transmission Mode

[0126] The user plane is an important component of the 5G network and is responsible for transmitting user data. It involves data transmission from the UE to the network and from the network to the UE.

[0127] The user plane interfaces are primarily the N3 and N9 interfaces. The N3 interface is the interface between the base station and the UPF, using the General Packet Radio Service (GPRS) Tunneling Protocol User Plane (GTP-U) protocol for tunneling user data. The N3 interface is primarily used to transmit uplink and downlink user plane data between the base station and the UPF. The N9 interface is the interface between different UPFs. In mobile scenarios, an intermediate UPF (I-UPF) can be inserted between the UE and the PDU Session Anchor (PSA) UPF for traffic forwarding. GTP-U is used between the two UPFs for user plane message transmission.

[0128] The GTP-U protocol is used to transmit user-plane data in mobile communication networks. It enables efficient data transmission between different network nodes by establishing tunnels and protocol encapsulation. The specific transmission mode used varies depending on the network version and network node.

[0129] In current 5G user-plane data transmission, the data transmission method between different nodes (such as a base station and UPF) is similar to the data transmission method shown in Figure 4, requiring significant CPU participation in the data transmission process. However, this data transmission scheme often encounters computer system bottlenecks caused by the mismatch between CPU and memory speeds, often referred to as the "memory wall." This is because CPU speeds are increasing, while memory speeds have not kept pace. Data transmission between the CPU and memory takes time. When the CPU needs to read or write large amounts of data, it waits for memory, resulting in the CPU being unable to fully utilize its computing power, affecting the overall performance of the computer system. This data transmission method, which requires significant CPU participation, is also known as messaging passing through the kernel. Because this data transmission method requires kernel transmission, it incurs high data movement and copying overhead. Furthermore, this data transmission method may not meet the high throughput and low latency requirements of certain services.

[0130] To reduce the CPU overhead of data transmission in 5G networks or future communication networks and meet business requirements for high throughput and low latency, RDMA transmission can be introduced into 5G networks or future communication networks. That is, RDMA links are established between two nodes in the 5G network or future communication network for data transmission.

[0131] 4. QoS Mapping Mechanism in 5G

[0132] QoS determines user satisfaction with the operator's services and is a comprehensive reflection of the network's service capabilities. QoS is required to provide different throughput rates and latency guarantees for users of different priorities within a cell. This ensures differentiation and fairness among multiple users, and provides services that match user requirements, ensuring that users' consumption reflects value.

[0133] QoS management is a network control mechanism that ensures that service quality requirements are met. It is an end-to-end process that requires the collaboration of all network nodes (UE <—> base station <—> core network) between the service initiator and responder to ensure service quality. In 5G standalone (SA) networking, the fundamental granularity of QoS management is the QoS flow. Service data flows (SDFs) are the basic units for implementing QoS management in the network. SDFs with the same QoS requirements and belonging to the same PDU session between the UE and the core network are called QoS flows. QoS flows are controlled by the SMF and can be pre-configured or established during the PDU session establishment or modification process. SDFs with the same QoS requirements are delivered through a single QoS flow, while SDFs with different QoS requirements are delivered through different QoS flows. Each QoS flow has an identifier, called a QoS flow identity (QFI). The QFI is simply an identifier and does not represent the QoS requirements of any particular SDF. Within a PDU session, each QoS flow has a unique QFI, meaning there is a one-to-one relationship between a QoS flow and a QFI. In SA networking, the QoS requirements of QoS flows are represented by 5G QoS identifiers (5QIs). Within the same PDU session, a 5QI can be used to represent a QoS requirement or the same QoS requirement for multiple QoS flows. Multiple QoS flows can be distinguished by the QFI.

[0134] In SA networking, IP flows are delivered to UEs via QoS flows. When a group of IP flows with the same service characteristics pass through the packet filter set (PFS) of the UPF, the PFS allocates an SDF with the same QoS requirements to these IP flows. The UPF maps multiple SDFs with the same QoS requirements and belonging to the same PDU session to the same QoS flow. The base station maps the QoS flow to the DRB and sends these IP flows to the UE via the DRB. Ultimately, these IP flows will be applied to the UE's APP. One DRB can correspond to one or more QoS flows.

[0135] In SA networking, the QoS requirements of QoS flows are no longer represented by QoS class identifiers (QCIs). Instead, they are represented by 5QIs. 5QIs are used to identify a set of 5G QoS characteristics, such as priority, packet latency, or packet error rate. In SA networking, the QoS requirements of a DRB actually represent the QoS requirements of the QoS flow to which it is mapped.

[0136] Figure 6 shows an example of QoS mapping in 5G. In this example, IP flow 1 is mapped to SDF1 by the UPF's PFS, IP flow 2 is mapped to SDF2 by the UPF's PFS, SDF1 and SDF2 are mapped to QoS flow 1, and QoS flow 1 is sent to the UE via DRB1. QoS flow 1 belongs to PDU session 1. IP flow 3 is mapped to SDF3 by the UPF's PFS, IP flows 4 and 5 are mapped to SDF4 by the UPF's PFS, SDF3 and SDF4 are mapped to QoS flow 2, and QoS flow 2 is sent to the UE via DRB2. QoS flow 2 belongs to PDU session 2. IP flow 6 is mapped to SDF5 by UPF's PFS, IP flow 7 is mapped to SDF6 by UPF's PFS, IP flow 8 is mapped to SDF7 by UPF's PFS, IP flow 9 and IP flow 10 are mapped to SDF8 by UPF's PFS, SDF5 and SDF6 are mapped to QoS flow 3, SDF7 and SDF8 are mapped to QoS flow 4, and QoS flow 3 and QoS flow 4 are sent to UE via DRB3, where QoS flow 3 and QoS flow 4 belong to PDU session 3.

[0137] 5. QoS Mapping Mechanism in RDMA

[0138] In RDMA transmission, a priority routing scheme is established, allowing each port to determine the priority of link traffic. Specifically, virtual lanes (VLs) are used to implement QoS. A virtual lane consists of a send buffer and a receive buffer, and a node requires at least two virtual lanes, specifically, VL0 and VL15. VL0 serves as a data buffer, and VL15 serves as a subnet manager packet (SMP) buffer. The SMP buffer can cache SMP packets carrying management device information. Optionally, a node may also include one or more of VL1-VL14, with VL1-VL14 serving as data buffers. Figure 7 shows an example of virtual lanes in RDMA. A port on node A is configured with VL0-VL1 and VL15, while a port on node B is configured with VL0-VL3 and VL15.

[0139] When implementing QoS functions, a node's port needs to rely on the Service Level to Virtual Lane Mapping Table (SLtoVLMappingTable). This mapping table can be set in the channel adapter (CA), switch, or router when the subnet manager (SM) configures the subnet. During transmission, each data packet contains an SL field, which is used to define the sending priority of the data packet required by the message sender. The sending priority is determined by the sender based on QoS requirements. When a data packet is sent to the link layer of the port for transmission, the link layer of the port determines which VL cache the data packet needs to be placed in based on the service level to virtual lane mapping table and the priority indicated by the SL field in the data packet. For example, if the priority indicated by the SL field in the data packet is x, the link layer of the port determines that the data packet needs to be placed in VL1 for transmission based on the service level and virtual lane mapping table.

[0140] As previously discussed, to reduce the CPU overhead of data transmission in 5G or future communication networks and meet service requirements for high throughput and low latency, RDMA transmission can be introduced into 5G or future communication networks. Specifically, RDMA links are established between two nodes in the 5G or future communication networks for data transmission. However, the QoS mapping mechanism in 5G or future communication networks differs from that in RDMA. Therefore, implementing RDMA QoS mapping when using RDMA transmission in 5G or future communication networks remains a challenge.

[0141] Figure 8(a) is a flow chart of a communication method provided in an embodiment of the present application. The method is executed by a first device or a module (such as a chip) of the first device, a second device or a module (such as a chip) of the second device, and a control plane network element or a module (such as a chip) of the control plane network element. The following description takes the execution of the method by the first device, the second device, and the control plane network element as an example. The first device is a UPF, the second device is a base station, or the first device is a base station and the second device is a UPF. The control plane network element is an SMF or a PCF, etc.

[0142] The method comprises the following steps:

[0143] Step 801a: The control plane network element sends first indication information to the first device. Correspondingly, the first device receives the first indication information.

[0144] The first indication information includes QP information of the first device, where the first indication information is used to indicate that a QP is established based on the QP information of the first device, where the QP information of the first device includes at least one of the following information: the number of QPs to be established, a memory region (MR) corresponding to the QP to be established, access rights of the QP to be established, a type of the QP to be established (such as RC, RD, UC, UD, or other types), a maximum transmission unit (MTU) of the QP to be established, a message length for transmission of the QP to be established, or a mapping relationship between the QP to be established and the QoS requirement.

[0145] The QoS requirement includes at least one of the following information: RDMA connection type (such as RC, RD, UC, UD, or other types), RDMA operation type (such as send, read, write), RDMA transmission type (such as lossy transmission or lossless transmission), RDMA protocol type (IB, RoCEv1, iWARP, RoCEv2, or other types of protocols), service latency requirement, service bandwidth requirement, service priority requirement, or transmission rate. For example, the QoS requirement can be represented by 5QI.

[0146] As an implementation method, the control plane network element may determine the QP information of the first device based on information about the first device, where the information about the first device may be reported by the first device to the control plane network element. The information about the first device includes at least one of hardware information of the first device, memory information of the first device, or RDMA information of the first device. The RDMA information of the first device includes an RDMA protocol supported by the first device and a bearer network type corresponding to the supported RDMA protocol.

[0147] Step 802a: The control plane network element sends second indication information to the second device. Correspondingly, the second device receives the second indication information.

[0148] The second indication information includes QP information of the second device, and the second indication information is used to indicate that a QP is established based on the QP information of the second device, where the QP information of the second device includes at least one of the following information: the number of QPs to be established, the memory area corresponding to the QP to be established, the access rights of the QP to be established, the type of the QP to be established, the maximum transmission unit of the QP to be established, the message length of the transmission of the QP to be established, or the mapping relationship between the QP to be established and the QoS requirement.

[0149] The QoS requirement includes at least one of the following information: RDMA connection type (such as RC, RD, UC, UD, or other types), RDMA operation type (such as send, read, write), RDMA transmission type (such as lossy transmission or lossless transmission), RDMA protocol type (IB, RoCEv1, iWARP, RoCEv2, or other types of protocols), service latency requirement, service bandwidth requirement, service priority requirement, or transmission rate. For example, the QoS requirement can be represented by 5QI.

[0150] As an implementation method, the control plane network element may determine the QP information of the second device based on information about the second device. The information about the second device may be reported to the control plane network element by the second device. The information about the second device includes at least one of hardware information of the second device, memory information of the second device, or RDMA information of the second device. The RDMA information of the second device includes an RDMA protocol supported by the second device and a bearer network type corresponding to the supported RDMA protocol.

[0151] It should be noted that there is no necessary relationship between the QP information of the first device and the QP information of the second device. For example, the number of QPs that the control plane network element instructs the first device to establish may be the same as or different from the number of QPs that the control plane network element instructs the second device to establish; the mapping relationship between the QP established by the first device and the QoS requirement may be the same as or different from the mapping relationship between the QP established by the control plane network element and the QoS requirement.

[0152] The order of step 801a and step 802a is not limited.

[0153] Step 803a: The first device establishes at least one QP according to the first indication information.

[0154] The first device establishes at least one QP based on the QP information of the first device indicated by the first indication information, and stores a mapping relationship between the at least one established QP and the QoS requirement. Information such as the number of QPs actually established by the first device, the MR corresponding to the QP, and the type of QP may be the same as or different from information such as the number of QPs to be established, the MR corresponding to the QP, and the type of QP indicated by the first indication information received by the first device, and this is not limited in this application.

[0155] For example, the first device establishes QP#1, QP#2 and QP#3, and the first indication information indicates that QP#1 and QP#2 both have a mapping relationship with 5QI#1, and QP#3 has a mapping relationship with 5QI#2. Therefore, the first device saves the mapping relationship between QP#1 and QP#2 and 5QI#1, and saves the mapping relationship between QP#3 and 5QI#2.

[0156] Exemplarily, after establishing at least one QP, the first device may further send QP establishment information of the first device to the control plane network element, where the QP establishment information of the first device is used to indicate information of the QP actually established by the first device.

[0157] Exemplarily, the control plane network element may update the QP information of the first device according to the QP establishment information of the first device, that is, the control plane network element updates the locally stored information related to the QP actually established by the first device.

[0158] Step 804a: The second device establishes at least one QP according to the second indication information.

[0159] The second device establishes at least one QP based on the QP information of the second device indicated by the second indication information, and stores a mapping relationship between the at least one established QP and the QoS requirement. Information such as the number of QPs actually established by the second device, the MR corresponding to the QP, and the type of QP may be the same as or different from the information such as the number of QPs to be established, the MR corresponding to the QP, and the type of QP indicated by the second indication information received by the second device, and this is not limited in this application.

[0160] For example, the second device establishes QP#a and QP#b, and the second indication information indicates that there is a mapping relationship between QP#a and 5QI#1, and there is a mapping relationship between QP#b and 5QI#2. Therefore, the second device saves the mapping relationship between QP#a and 5QI#1, and saves the mapping relationship between QP#b and 5QI#2.

[0161] Exemplarily, after establishing at least one QP, the second device may further send QP establishment information of the second device to the control plane network element, where the QP establishment information of the second device is used to indicate information of the QP actually established by the second device.

[0162] Exemplarily, the control plane network element may update the QP information of the second device according to the QP establishment information of the second device, that is, the control plane network element updates the locally stored relevant information about the QP actually established by the second device.

[0163] There is no restriction on the order between step 803a and step 802a and step 804a, and there is no restriction on the order between step 804a and step 801a.

[0164] As an implementation method, the above-mentioned first indication information is also used to indicate the establishment of an RDMA link between the first device and the second device, and the above-mentioned second indication information is also used to indicate the establishment of an RDMA link between the second device and the first device. After the above-mentioned step 804a, the following steps 805a to 806a can also be executed.

[0165] Step 805a: The first device establishes an RDMA link between the first device and the second device according to the established at least one QP.

[0166] Step 806a: The second device establishes an RDMA link between the first device and the second device according to the at least one established QP.

[0167] The order of step 805a and step 806a is not limited.

[0168] Among them, one or more RDMA links can be established between the first device and the second device. Each RDMA link corresponds to a QoS requirement. For example, the first device uses QP#1 and the second device uses QP#a, and RDMA link 1 is established between them. Also, the first device uses QP#3 and the second device uses QP#b, and RDMA link 2 is established between them. Among them, QP#1, QP#a and RDMA link 1 are all mapped to 5QI1, and QP#3, QP#b and RDMA link 2 are all mapped to 5QI2.

[0169] For example, the process of establishing an RDMA link between a first device and a second device is as follows: the first device sends the RDMA link establishment information of the first device to the second device, and the second device sends the RDMA link establishment information of the second device to the first device. Then, the first device and the second device establish an RDMA link based on the RDMA link establishment information of the first device and the RDMA link establishment information of the second device.

[0170] The RDMA link establishment information of the first device includes the port identifier and queue pair identifier of the first device. Optionally, the RDMA link establishment information of the first device also includes one or more of access permission information, source host address, destination host address, connection type, service level, protocol type, operation mode, or virtual memory address. Exemplarily, the port identifier of the first device includes a local identifier (LID), or includes a local identifier and a global identifier (GID). Exemplarily, the queue pair identifier can be a queue pair number (QPN). The port identifier and queue pair identifier of the first device can jointly indicate a QP address. Specifically, the port identifier indicates the port, and the queue pair identifier indicates the QP in the port. The source host address is the IP address of the first device. The destination host address is the IP address of the second device. The connection type is a reliable connection, an unreliable connection, a reliable datagram, an unreliable datagram, or other type. The service level is mainly applied to the QoS requirements of the RDMA service. Different service levels are used in RDMA to reflect the priority of different services. The protocol type is IB, RoCEv1, iWARP, RoCEv2, or other types of protocols. The operation mode is read, write or send, etc. The virtual memory address is the virtual address used by RDMA for communication in a work request, and the channel adapter can convert the virtual address into a physical address.

[0171] The RDMA link establishment information of the second device includes the port identifier and queue pair identifier of the second device. Optionally, the RDMA link establishment information of the second device also includes one or more of access permission information, source host address, destination host address, connection type, service level, protocol type, operation mode, or virtual memory address. Exemplarily, the port identifier of the second device includes a local identifier, or includes both a local identifier and a global identifier. Exemplarily, the queue pair identifier may be a queue pair number. The port identifier and queue pair identifier of the second device may jointly indicate a QP address. Specifically, the port identifier indicates the port, and the queue pair identifier indicates the QP in the port. The source host address is the IP address of the second device. The destination host address is the IP address of the first device. The connection type may be a reliable connection, an unreliable connection, a reliable datagram, an unreliable datagram, or another type. The service level is primarily applied to the QoS requirements of the RDMA service. Different service levels are used in RDMA to reflect the priority of different services. The protocol type may be IB, RoCEv1, iWARP, RoCEv2, or another type of protocol. The operation mode may be read, write, or send. The virtual memory address is the virtual address used by RDMA for communication in a work request. The channel adapter can translate the virtual address into a physical address.

[0172] After the RDMA link is established, the first device and the second device may select a corresponding RDMA link for data transmission based on the RDMA protocol.

[0173] As an implementation method, the control plane network element may indicate the QP to be used for data transmission, that is, the control plane network element may select the corresponding QP from the QPs that have been established in advance for service data transmission. For example, the control plane network element sends a notification message to the first device, the notification message including information about the first QP, the notification message being used to indicate the use of the first QP to transmit data packets of the first service, and sends a notification message to the second device, the notification message including information about the second QP, the notification message being used to indicate the use of the second QP to transmit data packets of the first service. Therefore, the first device uses the first QP to transmit data packets of the first service according to the notification message, and the second device uses the second QP to transmit data packets of the first service according to the notification message. Exemplarily, if the QoS requirement of the first service corresponds to 5QI1, the control plane network element may instruct the first device to use QP#1 to transmit data packets of the first service, and instruct the second device to use QP#a to transmit data packets of the first service, wherein both QP#1 and QP#a have been established in advance. Thus, the first device uses QP#1 to transmit data of the first service, and the second device uses QP#a to transmit data packets of the first service. That is, the first device and the second device transmit data packets of the first service based on RDMA link 1. The embodiment of FIG11 below is a specific example of this implementation method.

[0174] As another implementation method, the first device and the second device can autonomously select the QP to be used for data transmission, that is, the first device and the second device autonomously select the corresponding QP from the QPs that have been established in advance for service data transmission. For example, the control plane network element sends a packet detection rule (PDR) to the first device, and the PDR includes the service feature information of the first service, and the control plane network element sends a PDR to the second device, and the PDR includes the service feature information of the first service. The service feature information of the first service includes at least one of the following information: RDMA connection type, RDMA operation type, RDMA transmission type, RDMA protocol type, service latency requirement, service bandwidth requirement, service priority requirement or transmission rate. Then, the first device determines the first QoS requirement that matches the service feature information of the first service in the mapping relationship based on the mapping relationship between the locally stored QoS and QoS requirements, and determines the first QP corresponding to the first QoS requirement, and uses the first QP to transmit the data packet that matches the service feature information. The second device determines the first QoS requirement that matches the service characteristic information of the first service in the mapping relationship between the locally stored QoS and the QoS requirement, and determines the second QP corresponding to the first QoS requirement, and uses the second QP to transmit the data packet that matches the service characteristic information. For example, the service characteristic information of the first service corresponds to the first QoS requirement, and the first QoS requirement corresponds to 5QI1. The first device determines that the first QP corresponding to 5QI1 is QP#1, and the second device determines that the second QP corresponding to 5QI1 is QP#a, wherein QP#1 and QP#a have been established in advance. Thus, the first device uses QP#1 to transmit the data of the first service, and the second device uses QP#a to transmit the data packet of the first service, that is, the first device and the second device transmit the data packet of the first service based on RDMA link 1. The embodiment of Figure 12 below is a specific example of this implementation method.

[0175] As an implementation method, after the first device uses the first QP to transmit the data packet of the first service, it can also send feedback information to the control plane network element, where the feedback information is used to indicate the transmission status of the RDMA link corresponding to the first QP. Optionally, the control plane network element can send an update indication to the first device based on the feedback information, where the update indication is used to indicate the update of the QP to be used. For example, when the transmission quality of the RDMA link is lower than a certain quality threshold, the control plane network element can notify the first device to increase the memory size corresponding to the first QP, or the control plane network element can notify the first device to change the QP for data transmission, for example, from using QP#1 to transmit the data of the first service to using QP#2 to transmit the data of the first service.

[0176] As an implementation method, after the second device uses the second QP to transmit the data packet of the first service, it can also send feedback information to the control plane network element, and the feedback information is used to indicate the transmission status of the RDMA link corresponding to the second QP. Optionally, the control plane network element can send an update indication to the second device based on the feedback information, and the update indication is used to indicate the update of the QP to be used. For example, when the transmission quality of the RDMA link is lower than a certain quality threshold, the control plane network element can notify the second device to increase the memory size corresponding to the second QP, or the control plane network element can notify the second device to change the QP for data transmission, for example, from using QP#a to transmit the data of the first service to using QP#c to transmit the data of the first service.

[0177] The above solution proposes a device-level QP establishment process, in which the first and second devices pre-establish a device-level QP and RDMA link based on instructions from the control plane network element. After the RDMA link is pre-established, during the session establishment process, the first and second devices use one of the pre-established RDMA links that is adapted to the service's QoS requirements for service transmission. This allows the first and second devices to allocate data packets for services with different QoS requirements to corresponding QPs for transmission, thereby implementing QoS mapping when using RDMA transmission in 5G networks or future communication networks, thereby achieving QoS assurance. For example, after the RDMA link is pre-established, during the session establishment process, the control plane network element selects an appropriate QP for service transmission for the first and second devices based on the QoS requirements and the mapping relationship between the QP and the QoS requirements. This allows the first and second devices to allocate data packets for services with different QoS requirements to corresponding QPs for transmission, thereby achieving QoS mapping when using RDMA transmission in 5G networks or future communication networks, thereby achieving QoS assurance. For another example, when the RDMA link is established in advance, in the session establishment process, the control plane network element sends a PDR to the first device and the second device. The first device and the second device select a suitable QP for service transmission based on the PDR and the mapping relationship between QP and QoS requirements. In this way, the first device and the second device can allocate data packets of services with different QoS requirements to the corresponding QP for transmission, thereby realizing QoS mapping when using RDMA transmission in 5G networks or future communication networks, and further realizing QoS guarantee.

[0178] The embodiment of FIG. 9 below is a specific example of the embodiment of FIG. 8( a ).

[0179] Figure 8(b) is a flow chart of a communication method provided in an embodiment of the present application. The method is executed by a first device or a module (such as a chip) of the first device, a second device or a module (such as a chip) of the second device, and a control plane network element or a module (such as a chip) of the control plane network element. The following is an example of the first device, the second device, and the control plane network element executing the method. The first device is a UPF, the second device is a base station, or the first device is a base station and the second device is a UPF. The control plane network element is an SMF or a PCF, etc.

[0180] The method comprises the following steps:

[0181] Step 801b: The control plane network element sends first indication information to the first device. Correspondingly, the first device receives the first indication information.

[0182] The first indication information includes at least one QoS requirement, and the first indication information is used to instruct establishment of a QP that meets the at least one QoS requirement.

[0183] The QoS requirement includes at least one of the following information: RDMA connection type (such as RC, RD, UC, UD, or other types), RDMA operation type (such as send, read, write), RDMA transmission type (such as lossy transmission or lossless transmission), RDMA protocol type (IB, RoCEv1, iWARP, RoCEv2, or other types of protocols), service latency requirement, service bandwidth requirement, service priority requirement, or transmission rate. For example, the QoS requirement can be represented by 5QI.

[0184] As an implementation method, step 801b may specifically be: the control plane network element sends first indication information to the first device based on the information of the first device. The information of the first device includes at least one of hardware information of the first device, memory information of the first device, or RDMA information of the first device. The RDMA information of the first device includes an RDMA protocol supported by the first device and a bearer network type corresponding to the supported RDMA protocol.

[0185] Step 802b: The control plane network element sends second indication information to the second device. Correspondingly, the second device receives the second indication information.

[0186] The second indication information includes at least one QoS requirement, and the second indication information is used to instruct establishment of a QP that meets the at least one QoS requirement.

[0187] The QoS requirement includes at least one of the following information: RDMA connection type (such as RC, RD, UC, UD, or other types), RDMA operation type (such as send, read, write), RDMA transmission type (such as lossy transmission or lossless transmission), RDMA protocol type (IB, RoCEv1, iWARP, RoCEv2, or other types of protocols), service latency requirement, service bandwidth requirement, service priority requirement, or transmission rate. For example, the QoS requirement can be represented by 5QI.

[0188] As an implementation method, step 802b may specifically be: the control plane network element sends second indication information to the second device based on the information of the second device. The information of the second device includes at least one of hardware information of the second device, memory information of the second device, or RDMA information of the second device. The RDMA information of the second device includes an RDMA protocol supported by the second device and a bearer network type corresponding to the supported RDMA protocol.

[0189] The order of step 801b and step 802b is not limited.

[0190] Step 803b: The first device establishes at least one QP that meets at least one QoS requirement according to the first indication information, and establishes a mapping relationship between the at least one QoS requirement and the at least one QP.

[0191] For example, the first device establishes QP#1, QP#2, and QP#3, and establishes a mapping relationship between QP#1 and QP#2 and 5QI#1, and a mapping relationship between QP#3 and 5QI#2.

[0192] Exemplarily, after establishing at least one QP, the first device may further send QP establishment information of the first device to the control plane network element, where the QP establishment information of the first device is used to indicate information of the QP actually established by the first device.

[0193] Exemplarily, the control plane network element may update the QP information of the first device according to the QP establishment information of the first device, that is, the control plane network element updates the locally stored information related to the QP actually established by the first device.

[0194] Step 804b: The second device establishes at least one QP that meets the at least one QoS requirement according to the second indication information, and establishes a mapping relationship between the at least one QoS requirement and the at least one QP.

[0195] For example, the second device establishes QP#a and QP#b, and establishes a mapping relationship between QP#a and 5QI#1, and a mapping relationship between QP#b and 5QI#2.

[0196] Exemplarily, after establishing at least one QP, the second device may further send QP establishment information of the second device to the control plane network element, where the QP establishment information of the second device is used to indicate information of the QP actually established by the second device.

[0197] Exemplarily, the control plane network element may update the QP information of the second device according to the QP establishment information of the second device, that is, the control plane network element updates the locally stored relevant information about the QP actually established by the second device.

[0198] There is no restriction on the order between step 803b and step 802b and step 804b, and there is no restriction on the order between step 804b and step 801b.

[0199] As an implementation method, the above-mentioned first indication information is also used to indicate the establishment of an RDMA link between the first device and the second device, and the above-mentioned second indication information is also used to indicate the establishment of an RDMA link between the second device and the first device. After the above-mentioned step 804a, the following steps 805b to 806b can also be executed.

[0200] Step 805b: The first device establishes an RDMA link between the first device and the second device according to the at least one established QP.

[0201] Step 806b: The second device establishes an RDMA link between the first device and the second device according to the at least one established QP.

[0202] The order of step 805b and step 806b is not limited.

[0203] Among them, one or more RDMA links can be established between the first device and the second device. Each RDMA link corresponds to a QoS requirement. For example, the first device uses QP#1 and the second device uses QP#a, and RDMA link 1 is established between them. Also, the first device uses QP#3 and the second device uses QP#b, and RDMA link 2 is established between them. Among them, QP#1, QP#a and RDMA link 1 are all mapped to 5QI1, and QP#3, QP#b and RDMA link 2 are all mapped to 5QI2.

[0204] For example, the process of establishing an RDMA link between a first device and a second device is as follows: the first device sends its RDMA link establishment information to the second device, and the second device sends its RDMA link establishment information to the first device. Then, the first and second devices establish an RDMA link based on the RDMA link establishment information of the first device and the RDMA link establishment information of the second device. For details about the RDMA link establishment information of the first device and the RDMA link establishment information of the second device, please refer to the description of the embodiment of Figure 8(a) above and will not be repeated here.

[0205] After the RDMA link is established, the first device and the second device may select a corresponding RDMA link for data transmission based on the RDMA protocol.

[0206] As an implementation method, the control plane network element may indicate the QP to be used for data transmission, that is, the control plane network element may select the corresponding QP from the QPs that have been established in advance for service data transmission. For example, the control plane network element sends a notification message to the first device, the notification message including information about the first QP, the notification message being used to indicate the use of the first QP to transmit data packets of the first service, and sends a notification message to the second device, the notification message including information about the second QP, the notification message being used to indicate the use of the second QP to transmit data packets of the first service. Therefore, the first device uses the first QP to transmit data packets of the first service according to the notification message, and the second device uses the second QP to transmit data packets of the first service according to the notification message. Exemplarily, if the QoS requirement of the first service corresponds to 5QI1, the control plane network element may instruct the first device to use QP#1 to transmit data packets of the first service, and instruct the second device to use QP#a to transmit data packets of the first service, wherein both QP#1 and QP#a have been established in advance. Thus, the first device uses QP#1 to transmit data of the first service, and the second device uses QP#a to transmit data packets of the first service. That is, the first device and the second device transmit data packets of the first service based on RDMA link 1. The embodiment of FIG11 below is a specific example of this implementation method.

[0207] As another implementation method, the first device and the second device can autonomously select the QP to use for data transmission. Specifically, the first device and the second device autonomously select the corresponding QP from pre-established QPs for service data transmission. For example, the control plane network element sends a PDR to the first device, where the PDR includes service feature information for the first service, and the control plane network element sends a PDR to the second device, where the PDR includes service feature information for the first service. The service feature information for the first service includes at least one of the following: RDMA connection type, RDMA operation type, RDMA transmission type, RDMA protocol type, service latency requirement, service bandwidth requirement, service priority requirement, or transmission rate. The first device then determines, based on a locally stored mapping relationship between QoS and QoS requirements, a first QoS requirement that matches the service feature information for the first service, determines a first QP corresponding to the first QoS requirement, and uses the first QP to transmit data packets that match the service feature information. The second device determines, based on a locally stored mapping relationship between QoS and QoS requirements, a first QoS requirement that matches the service feature information for the first service, determines a second QP corresponding to the first QoS requirement, and uses the second QP to transmit data packets that match the service feature information. For example, the service characteristic information of the first service corresponds to a first QoS requirement, which corresponds to 5QI1. The first device determines that the first QP corresponding to 5QI1 is QP#1, and the second device determines that the second QP corresponding to 5QI1 is QP#a, where both QP#1 and QP#a have been established in advance. Thus, the first device uses QP#1 to transmit data for the first service, and the second device uses QP#a to transmit data packets for the first service. That is, the first and second devices transmit data packets for the first service over RDMA link 1. The embodiment of FIG12 below is a specific example of this implementation method.

[0208] As an implementation method, after the first device uses the first QP to transmit the data packet of the first service, it can also send feedback information to the control plane network element, where the feedback information is used to indicate the transmission status of the RDMA link corresponding to the first QP. Optionally, the control plane network element can send an update indication to the first device based on the feedback information, where the update indication is used to indicate the update of the QP to be used. For example, when the transmission quality of the RDMA link is lower than a certain quality threshold, the control plane network element can notify the first device to increase the memory size corresponding to the first QP, or the control plane network element can notify the first device to change the QP for data transmission, for example, from using QP#1 to transmit the data of the first service to using QP#2 to transmit the data of the first service.

[0209] As an implementation method, after the second device uses the second QP to transmit the data packet of the first service, it can also send feedback information to the control plane network element, and the feedback information is used to indicate the transmission status of the RDMA link corresponding to the second QP. Optionally, the control plane network element can send an update indication to the second device based on the feedback information, and the update indication is used to indicate the update of the QP to be used. For example, when the transmission quality of the RDMA link is lower than a certain quality threshold, the control plane network element can notify the second device to increase the memory size corresponding to the second QP, or the control plane network element can notify the second device to change the QP for data transmission, for example, from using QP#a to transmit the data of the first service to using QP#c to transmit the data of the first service.

[0210] The above solution proposes a device-level QP establishment process, in which the first and second devices independently negotiate to establish a QP and establish an RDMA link based on the QoS requirements provided by the control plane network element. After the RDMA link is pre-established, during the session establishment process, the first and second devices use a pre-established RDMA link that is adapted to the service's QoS requirements for service transmission. This allows the first and second devices to allocate data packets for services with different QoS requirements to corresponding QPs for transmission, thus implementing QoS mapping when using RDMA transmission in 5G networks or future communication networks, thereby achieving QoS assurance. For example, after the RDMA link is pre-established, during the session establishment process, the control plane network element selects an appropriate QP for service transmission for the first and second devices based on the QoS requirements and the mapping relationship between QPs and QoS requirements. This allows the first and second devices to allocate data packets for services with different QoS requirements to corresponding QPs for transmission, thus implementing QoS mapping when using RDMA transmission in 5G networks or future communication networks, thereby achieving QoS assurance. For another example, when the RDMA link is established in advance, in the session establishment process, the control plane network element sends a PDR to the first device and the second device. The first device and the second device select a suitable QP for service transmission based on the PDR and the mapping relationship between QP and QoS requirements. In this way, the first device and the second device can allocate data packets of services with different QoS requirements to the corresponding QP for transmission, thereby realizing QoS mapping when using RDMA transmission in 5G networks or future communication networks, and further realizing QoS guarantee.

[0211] The embodiment of FIG10 below is a specific example of the embodiment of FIG8( b ).

[0212] The embodiments of FIG. 8( a ) and FIG. 8( b ) are described below with reference to the specific examples of FIG. 9 to FIG. 12 .

[0213] Figure 9 is a flow chart of a communication method provided in an embodiment of the present application. In this embodiment, a control plane network element (such as an SMF or PCF) indicates information about a QP to be established to a base station and a UPF, and the base station and the UPF establish the QP based on the instruction of the control plane network element. The method includes the following steps:

[0214] Step 901: AMF sends base station information to SMF. Correspondingly, SMF receives the base station information.

[0215] The information of the base station includes at least one of hardware information of the base station, memory information of the base station, or RDMA information of the base station.

[0216] The hardware information includes network card information and / or CPU information. The network card information includes at least one of the network card model, driver information, or supported transmission rate.

[0217] The memory information includes at least one of a memory model, a memory size, or a memory usage.

[0218] RDMA information includes supported RDMA protocols (such as RoceV1, RoceV2, Iwarp, IB or other protocol types) and the bearer network type corresponding to the supported RDMA protocol.

[0219] Step 902: The UPF sends the UPF information to the SMF. Correspondingly, the SMF receives the UPF information.

[0220] The information of the UPF includes at least one of hardware information of the UPF, memory information of the UPF, or RDMA information of the UPF.

[0221] The hardware information includes network card information and / or CPU information. The network card information includes at least one of the network card model, driver information, or supported transmission rate.

[0222] The memory information includes at least one of a memory model, a memory size, or a memory usage.

[0223] RDMA information includes supported RDMA protocols (such as RoceV1, RoceV2, Iwarp, IB or other protocol types) and the bearer network type corresponding to the supported RDMA protocol.

[0224] The order of the above steps 901 and 902 is not limited.

[0225] Step 903: The SMF or PCF determines the RDMA link establishment strategy based on the information of the base station and the information of the UPF.

[0226] If the RDMA link establishment strategy is determined by the PCF, then before step 903, the SMF also sends the base station information and the UPF information to the PCF.

[0227] The RDMA link establishment strategy includes the QP information of the base station and the QP information of the UPF.

[0228] The QP information of the base station includes at least one of the following information: the number of QPs to be established, the memory area corresponding to the QP to be established, the access rights of the QP to be established, the type of the QP to be established (such as RC, RD, UC, UD or other types), the maximum transmission unit of the QP to be established, the message length of the transmission of the QP to be established, or the mapping relationship between the QP to be established and the QoS requirement. Among them, the QoS requirement includes at least one of the following information: RDMA connection type (such as RC, RD, UC, UD or other types), RDMA operation type (such as send, read, write), RDMA transmission type (such as lossy transmission or lossless transmission), RDMA protocol type (IB, RoCEv1, iWARP, RoCEv2 or other types of protocols), service delay requirement, service bandwidth requirement, service priority requirement or transmission rate. For example, the QoS requirement can be represented by 5QI.

[0229] The QP information of the UPF includes at least one of the following information: the number of QPs to be established, the MR corresponding to the QP to be established, the access rights of the QP to be established, the type of the QP to be established (such as RC, RD, UC, UD or other types), the MTU of the QP to be established, the message length of the transmission of the QP to be established, or the mapping relationship between the QP to be established and the QoS requirement. Among them, the QoS requirement includes at least one of the following information: RDMA connection type (such as RC, RD, UC, UD or other types), RDMA operation type (such as send, read, write), RDMA transmission type (such as lossy transmission or lossless transmission), RDMA protocol type (IB, RoCEv1, iWARP, RoCEv2 or other types of protocols), service latency requirements, service bandwidth requirements, service priority requirements or transmission rate. For example, the QoS requirement can be represented by 5QI.

[0230] Step 904: The SMF sends an indication message to the UPF, and the UPF receives the indication message accordingly.

[0231] If the PCF determines the RDMA link establishment policy in step 903, the PCF first sends the indication information to the SMF, which then sends the indication information to the UPF. Alternatively, the PCF sends the UPF's QP information to the SMF, and the SMF generates the indication information based on the UPF's QP information.

[0232] If the RDMA link establishment policy is generated by the SMF in the above step 903, the SMF generates indication information according to the QP information of the UPF.

[0233] The indication information includes the QP information of the UPF, and is used to instruct the establishment of a QP based on the QP information of the UPF, and optionally also instructs the UPF to establish an RDMA link with the base station.

[0234] Step 905: The SMF sends indication information to the base station. Correspondingly, the base station receives the indication information.

[0235] Step 905 is specifically as follows: SMF sends indication information to AMF, and then AMF forwards the indication information to the base station.

[0236] If the PCF determines the RDMA link establishment strategy in step 903, the PCF first sends the indication information to the SMF, which then sends the indication information to the base station. Alternatively, the PCF sends the base station's QP information to the SMF, and the SMF generates the indication information based on the base station's QP information.

[0237] If the RDMA link establishment strategy is generated by the SMF in the above step 903, the SMF generates indication information according to the QP information of the base station.

[0238] The indication information includes the QP information of the base station, and is used to instruct the establishment of a QP based on the QP information of the base station, and optionally also instructs the base station to establish an RDMA link with the UPF.

[0239] The order of step 904 and step 905 is not limited.

[0240] Step 906: The UPF establishes an RDMA link with the base station.

[0241] The UPF establishes a corresponding QP according to the received indication information, and the base station establishes a corresponding QP according to the received indication information. Then, an RDMA link is established between the UPF and the base station based on the established QP.

[0242] It should be noted that the number of QPs actually established by the UPF, the MR corresponding to the QP, the type of QP, and other information may be the same as or different from the number of QPs to be established, the MR corresponding to the QP, the type of QP, and other information in the indication information received by the UPF. This application does not limit this. Similarly, the number of QPs actually established by the base station, the MR corresponding to the QP, the type of QP, and other information may be the same as or different from the number of QPs to be established, the MR corresponding to the QP, the type of QP, and other information in the indication information received by the base station. This application does not limit this.

[0243] In step 907, the UPF sends the QP establishment information of the UPF to the SMF. Correspondingly, the SMF receives the QP establishment information of the UPF.

[0244] The QP establishment information is used to indicate the relevant information of the QP actually established by the UPF, such as the number of QPs actually established by the UPF, the MR corresponding to the QP, the type of QP, etc. For example, the UPF establishes QP#a and QP#b, and QP#a is mapped to 5QI#1, and QP#b is mapped to 5QI#2.

[0245] In step 908, the base station sends the QP establishment information of the base station to the SMF. Correspondingly, the SMF receives the QP establishment information of the base station.

[0246] Step 908 is specifically as follows: the base station sends the QP establishment information of the base station to the AMF, and then the AMF forwards the QP establishment information of the base station to the SMF.

[0247] The QP establishment information is used to indicate the relevant information of the QP actually established by the base station, such as the number of QPs actually established by the base station, the MR corresponding to the QP, the type of QP, etc. For example, the base station establishes QP#1, QP#2, and QP#3, and QP#1 and QP#2 are mapped to 5QI#1, and QP#3 is mapped to 5QI#2.

[0248] Optionally, if the above step 903 is performed by the PCF, the SMF also sends the QP establishment information of the UPF and the QP establishment information of the base station to the PCF.

[0249] The order of step 907 and step 908 is not limited.

[0250] The above solution proposes a device-level QP establishment process. The base station and UPF establish a QP and establish an RDMA link based on instructions from the control plane's SMF or PCF. Later in the session process, the base station and UPF can allocate data packets for services with different QoS requirements to the corresponding QP for transmission based on the established QP and the mapping between QP and QoS requirements. This implements QoS mapping when using RDMA transmission in 5G networks and future communication networks, thereby achieving QoS assurance.

[0251] FIG10 is a flow chart of a communication method provided in an embodiment of the present application. In this embodiment, the base station and the UPF negotiate and independently decide to establish a QP. The method includes the following steps:

[0252] Steps 1001 to 1002 are the same as steps 901 to 902 in the embodiment of FIG. 9 .

[0253] Step 1003: The SMF or PCF sends an indication message to the UPF based on the UPF information. Correspondingly, the UPF receives the indication message.

[0254] Among them, if the indication information is sent by PCF to UPF, before step 1003, SMF also sends UPF information to PCF.

[0255] The indication information includes at least one QoS requirement, and the indication information is used to instruct the UPF to establish a QP that meets the at least one QoS requirement and establish an RDMA link between the UPF and the base station. The QoS requirement includes at least one of the following information: RDMA connection type (such as RC, RD, UC, UD or other types), RDMA operation type (such as send, read, write), RDMA transmission type (such as lossy transmission or lossless transmission), RDMA protocol type (IB, RoCEv1, iWARP, RoCEv2 or other types of protocols), service delay requirements, service bandwidth requirements, service priority requirements or transmission rate. For example, the QoS requirement can be represented by 5QI.

[0256] Specifically, the SMF or PCF receives information from the UPF, determines the QoS requirements that the UPF can meet, and then carries the QoS requirements in the indication information and sends it to the UPF.

[0257] Step 1004: The SMF or PCF sends indication information to the base station. Correspondingly, the base station receives the indication information.

[0258] If the SMF sends the indication information to the base station, it can be specifically: SMF sends the indication information to AMF, and then AMF forwards the indication information to the base station. If the PCF sends the indication information to the base station, it can be specifically: PCF sends the indication information to SMF, SMF sends the indication information to AMF, and then AMF forwards the indication information to the base station.

[0259] The indication information includes at least one QoS requirement, and the indication information is used to instruct the base station to establish a QP that meets the at least one QoS requirement and establish an RDMA link between the base station and the UPF. The QoS requirement includes at least one of the following information: RDMA connection type (such as RC, RD, UC, UD or other types), RDMA operation type (such as send, read, write), RDMA transmission type (such as lossy transmission or lossless transmission), RDMA protocol type (IB, RoCEv1, iWARP, RoCEv2 or other types of protocols), service latency requirements, service bandwidth requirements, service priority requirements or transmission rate. For example, the QoS requirement can be represented by 5QI.

[0260] Specifically, the SMF or PCF receives information from the base station, determines the QoS requirements that the base station can meet, and then carries the QoS requirements in the indication information and sends it to the base station.

[0261] The order of step 1003 and step 1004 is not limited.

[0262] Step 1005: The UPF establishes an RDMA link with the base station.

[0263] Based on the received indication information, the UPF establishes a QP that meets the QoS requirements and establishes a mapping relationship between the QoS requirements and the QP. For example, the UPF establishes QP#a and QP#b, and QP#a is mapped to 5QI#1, and QP#b is mapped to 5QI#2.

[0264] Based on the received indication information, the base station establishes a QP that meets the QoS requirements and establishes a mapping relationship between the QoS requirements and the QP. For example, the base station establishes QP#1, QP#2, and QP#3, and QP#1 and QP#2 are mapped to 5QI#1, and QP#3 is mapped to 5QI#2.

[0265] After the QP is established, an RDMA link is established between the UPF and the base station based on the established QP.

[0266] Step 1006: The UPF sends the UPF's QP establishment information to the SMF. Correspondingly, the SMF receives the UPF's QP establishment information.

[0267] The QP establishment information is used to indicate the relevant information of the QP established by the UPF and the mapping relationship between the established QP and the QoS requirements. The relevant information of the established QP includes, for example, the number of QPs actually established by the UPF, the MR corresponding to the QP, the type of QP, and other information.

[0268] Step 1007: The base station sends the QP establishment information of the base station to the SMF. Correspondingly, the SMF receives the QP establishment information of the base station.

[0269] Step 1007 is specifically as follows: the base station sends the QP establishment information of the base station to the AMF, and then the AMF forwards the QP establishment information of the base station to the SMF.

[0270] The QP establishment information is used to indicate the relevant information of the QP established by the base station and the mapping relationship between the established QP and the QoS requirement. The relevant information of the established QP includes, for example, the number of QPs actually established by the base station, the MR corresponding to the QP, the type of QP, and other information.

[0271] Optionally, if the above steps 1003 and 1004 are performed by the PCF, the SMF also sends the QP establishment information of the UPF and the QP establishment information of the base station to the PCF.

[0272] The order of step 1006 and step 1007 is not limited.

[0273] The above solution proposes a device-level QP establishment process. The base station and UPF independently negotiate to establish a QP and establish an RDMA link based on the QoS requirements provided by the control plane SMF or PCF. Later in the session process, the base station and UPF can allocate data packets of services with different QoS requirements to the corresponding QP for transmission based on the established QP and the mapping between QP and QoS requirements. This implements QoS mapping when using RDMA transmission in 5G networks and future communication networks, thereby achieving QoS assurance.

[0274] Figure 11 is a flow chart of a communication method provided in an embodiment of the present application. In this embodiment, during the session establishment process, a control plane network element (such as an SMF or PCF) allocates an appropriate QP for service transmission. The method includes the following steps:

[0275] Step 1101: Complete the RDMA link establishment between the base station and the UPF.

[0276] This step 1101 can be completed by the embodiment of Figure 9, that is, this step 1101 can be replaced by steps 901 to 908 of the embodiment of Figure 9. Alternatively, this step 1101 can be completed by the embodiment of Figure 10, that is, this step 1101 can be replaced by steps 1001 to 1007 of the embodiment of Figure 10.

[0277] Step 1102: The UE sends a PDU session establishment request to the AMF. Accordingly, the AMF receives the PDU session establishment request.

[0278] The PDU session establishment request is used to request establishment of a PDU session.

[0279] Exemplarily, the PDU session establishment request may be a PDU_Session_Establishment Request.

[0280] Step 1103: AMF sends a PDU session context establishment request to SMF. Correspondingly, SMF receives the PDU session context establishment request.

[0281] The PDU session context establishment request is used to request establishment of a PDU session context.

[0282] Exemplarily, the PDU session establishment request may be Nsmf_PDUSession_CreateSMContext Request.

[0283] Step 1104: SMF or PCF determines to adopt RDMA transmission.

[0284] In one implementation method, the SMF or PCF receives a request message from the UE, where the request message is used to request the establishment of an RDMA link, and determines to adopt RDMA transmission based on the request message.

[0285] In another implementation method, the SMF or PCF receives a QoS request including a QoS level, and the data transmission mode corresponding to the QoS level is RDMA transmission. The SMF or PCF determines to use RDMA transmission based on the QoS request.

[0286] In another implementation method, the SMF or PCF determines to adopt RDMA transmission according to local configuration information.

[0287] Step 1105: The SMF or PCF allocates a QP to the UPF according to the QoS requirements of the service and the mapping relationship between the QP of the UPF and the QoS requirements.

[0288] In one implementation method, if the above step 1101 is implemented based on the embodiment of FIG. 9 , the mapping relationship between the QP of the UPF and the QoS requirement is generated by the SMF or the PCF.

[0289] In another implementation method, if the above step 1101 is implemented based on the embodiment of Figure 10, the mapping relationship between the QP of the UPF and the QoS requirement is generated by the UPF and sent to the SMF or PCF.

[0290] Step 1106: The SMF sends the information of the QP allocated to the UPF to the UPF. Correspondingly, the UPF receives the information of the QP allocated to the UPF.

[0291] The information of the QP allocated to the UPF may include the GID and QPN, or include the GID, LID and QPN. That is, the information of the QP allocated to the UPF uniquely identifies a QP.

[0292] In step 1107, the SMF or PCF allocates a QP to the base station according to the QoS requirement of the service and the mapping relationship between the QP of the base station and the QoS requirement.

[0293] In one implementation method, if the above step 1101 is implemented based on the embodiment of FIG. 9 , the mapping relationship between the QP of the base station and the QoS requirement is generated by the SMF or PCF.

[0294] In another implementation method, if the above step 1101 is implemented based on the embodiment of Figure 10, the mapping relationship between the QP of the base station and the QoS requirement is generated by the base station and sent to the SMF or PCF.

[0295] Step 1108: The SMF sends information about the QP allocated to the base station to the base station. Correspondingly, the base station receives information about the QP allocated to the base station.

[0296] The information of the QP allocated to the base station may include the GID and QPN, or include the GID, LID and QPN. That is, the information of the QP allocated to the base station uniquely identifies a QP.

[0297] It should be noted that there is no restriction on the order of execution of steps 1105 and 1106, and steps 1107 and 1108. Specifically, there is no restriction on the order of execution of steps 1105 and 1107, there is no restriction on the order of execution of steps 1105 and 1108, there is no restriction on the order of execution of steps 1106 and 1107, and there is no restriction on the order of execution of steps 1106 and 1108.

[0298] It should be noted that steps 1105 and 1107 can also be combined into a single step. For example, the SMF or PCF allocates a QP to the UPF based on the service's QoS requirements and the mapping between the UPF's QP and QoS requirements. After allocating the QP to the UPF, if an RDMA link has already been established between the QP and one or more QPs of the base station, the SMF or PCF directly allocates a QP to the base station based on the RDMA link. For example, if QP#a is allocated to the UPF, and an RDMA link has already been established between QP#1 of the base station, the SMF or PCF determines that the QP allocated to the base station is QP#1. For another example, the SMF can first allocate a QP to the base station based on the service's QoS requirements and the mapping between the base station's QP and QoS requirements. Then, after allocating the QP to the base station, if an RDMA link has already been established between the QP and one or more QPs of the UPF, the SMF or PCF directly allocates a QP to the UPF based on the RDMA link. For example, if QP#1 is allocated to the base station and an RDMA link has been established between QP#1 and QP#a of the UPF in advance, the SMF or PCF determines that the QP allocated to the UPF is QP#a.

[0299] Step 1109, the remaining process of PDU session establishment.

[0300] Step 1110: The UPF and the base station use the allocated QP to perform data transmission.

[0301] For example, if the SMF or PCF allocates QP#a to the UPF and QP#1 to the base station, the UPF uses the allocated QP as QP#a and the base station uses the allocated QP as QP#1, and both parties perform service transmission.

[0302] In step 1111, the base station sends the transmission status of the RDMA link to the SMF. Correspondingly, the SMF receives the transmission status of the RDMA link.

[0303] The specific step 1111 is: the base station sends the transmission status of the RDMA link to the AMF, and then the AMF forwards the transmission status of the RDMA link to the SMF.

[0304] That is, when the base station uses the QP allocated by the SMF or PCF to transmit services, it also monitors the transmission status of the RDMA link and reports the transmission status of the RDMA link to the SMF.

[0305] Exemplarily, the transmission status of the RDMA link includes the memory usage corresponding to the QP (such as the size of the remaining allocatable memory), the amount of data to be transmitted, bandwidth latency, throughput, etc.

[0306] In step 1112, the UPF sends the transmission status of the RDMA link to the SMF. Correspondingly, the SMF receives the transmission status of the RDMA link.

[0307] That is, when the UPF uses the QP allocated by the SMF or PCF for business transmission, it also monitors the transmission status of the RDMA link and reports the transmission status of the RDMA link to the SMF.

[0308] Exemplarily, the transmission status of the RDMA link includes the memory usage corresponding to the QP (such as the size of the remaining allocatable memory), the amount of data to be transmitted, bandwidth latency, throughput, etc.

[0309] Step 1113: The SMF or PCF adjusts the QP according to the transmission status of the RDMA link.

[0310] The SMF or PCF determines whether the currently allocated QP can meet the business needs or whether the transmission quality meets the requirements based on the transmission status of the RDMA link reported by the base station and / or the transmission status of the RDMA link reported by the UPF. If not, the parameters of the currently allocated QP can be adjusted, such as the memory size corresponding to the QP, the rate of the QP, the bandwidth of the QP, the transmission priority of the QP, or the packet sending interval.

[0311] The above steps 1111 to 1113 are optional steps.

[0312] In the above solution, when the RDMA link is established in advance, in the session establishment process, the control plane network element selects the appropriate QP for the base station and UPF for service transmission based on the QoS requirements and the mapping relationship between QP and QoS requirements. In this way, the base station and UPF can allocate data packets of services with different QoS requirements to the corresponding QP for transmission, realizing QoS mapping when using RDMA transmission in 5G networks or future communication networks, and thus achieving QoS guarantee.

[0313] Figure 12 is a flow chart of a communication method provided in an embodiment of the present application. In this embodiment, during the session establishment process, the base station and the UPF independently select an appropriate QP for service transmission. The method includes the following steps:

[0314] Steps 1201 to 1204 are the same as steps 1101 to 1104 in the embodiment of FIG. 11 .

[0315] Step 1205: The SMF or PCF generates a PDR.

[0316] The PDR contains service feature information, which includes at least one of the following information: RDMA connection type (such as RC, RD, UC, UD or other types), RDMA operation type (such as send, read, write), RDMA transmission type (such as lossy transmission or lossless transmission), RDMA protocol type (IB, RoCEv1, iWARP, RoCEv2 or other types of protocols), service latency requirements, service bandwidth requirements, service priority requirements or transmission rate.

[0317] Step 1206: The SMF sends a PDR to the UPF. Correspondingly, the UPF receives the PDR.

[0318] Step 1207: The SMF sends a PDR to the base station. Correspondingly, the base station receives the PDR.

[0319] The specific step 1207 is: SMF sends PDR to AMF, and then AMF forwards the PDR to the base station.

[0320] Step 1208, the remaining process of PDU session establishment.

[0321] Step 1209: UPF selects an appropriate QP based on the PDR.

[0322] The UPF determines the corresponding QoS requirements based on the service feature information in the PDR, and then selects the corresponding QP based on the mapping relationship between QP and QoS requirements.

[0323] Step 1210: The base station selects an appropriate QP according to the PDR.

[0324] The base station determines the corresponding QoS requirements based on the service feature information in the PDR, and then selects the corresponding QP based on the mapping relationship between QP and QoS requirements.

[0325] The order of step 1209 and step 1210 is not limited.

[0326] Step 1211: The UPF and the base station use the selected QP for data transmission.

[0327] For example, if the QP selected by the UPF is QP#a and the QP selected by the base station is QP#1, then the QP used by the UPF is QP#a and the QP used by the base station is QP#1, and both parties perform service transmission.

[0328] Steps 1212 to 1214 are the same as steps 1111 to 1113 in the embodiment of FIG. 11 .

[0329] The above steps 1212 to 1214 are optional steps.

[0330] In the above solution, when the RDMA link is established in advance, in the session establishment process, the control plane network element sends the PDR to the base station and UPF. The UPF and the base station select the appropriate QP for service transmission based on the mapping relationship between the PDR and the QP and the QoS requirements. In this way, the base station and UPF can allocate data packets of services with different QoS requirements to the corresponding QP for transmission, realizing QoS mapping when using RDMA transmission in 5G networks or future communication networks, and thus achieving QoS guarantee.

[0331] It is understood that, in order to implement the functions in the above embodiments, the first device, the second device, or the control plane network element includes hardware structures and / or software modules corresponding to the execution of each function. Those skilled in the art should readily appreciate that, in combination with the units and method steps of the various examples described in the embodiments disclosed in this application, this application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in hardware or in a manner driven by computer software depends on the specific application scenario and design constraints of the technical solution.

[0332] Figures 13 and 14 are schematic diagrams of the structures of the communication devices provided in the embodiments of the present application. These communication devices can be used to implement the functions of the first device, the second device, or the control plane network element in the above method embodiments, and thus can also achieve the beneficial effects of the above method embodiments. In the embodiments of the present application, the communication device can be the first device, the second device, or the control plane network element, or it can be a module (such as a chip) applied to the first device, the second device, or the control plane network element.

[0333] The communication device 1300 shown in Figure 13 includes a processing unit 1310 and a transceiver unit 1320. The communication device 1300 is used to implement the functions of the first device, the second device or the control plane network element in the above method embodiment.

[0334] When the communication device 1300 is used to implement the function of the first device in the above method embodiment, the transceiver unit 1320 is configured to receive indication information from a control plane network element, where the indication information includes QP information of the first device, and the indication information is used to indicate establishment of a QP based on the QP information of the first device, where the QP information of the first device includes at least one of the following information: the number of QPs to be established, the memory area corresponding to the QP to be established, the access rights of the QP to be established, the type of the QP to be established, the maximum transmission unit of the QP to be established, the message length of the transmission of the QP to be established, or a mapping relationship between the QP to be established and the QoS requirement; and the processing unit 1310 is configured to establish at least one QP according to the indication information.

[0335] In a possible implementation method, the indication information is further used to instruct establishment of an RDMA link between the first device and the second device; and the processing unit 1310 is further used to establish the RDMA link between the first device and the second device according to the at least one QP.

[0336] In one possible implementation method, the QoS requirement includes at least one of the following information: RDMA connection type, RDMA operation type, RDMA transmission type, RDMA protocol type, service latency requirement, service bandwidth requirement, service priority requirement or transmission rate.

[0337] In a possible implementation method, the transceiver unit 1320 is further configured to send QP establishment information of the first device to the control plane network element, where the QP establishment information of the first device is used to indicate information of the QP actually established by the first device.

[0338] In one possible implementation method, the transceiver unit 1320 is further used to receive a PDR from the control plane network element, where the PDR includes service feature information of the first service, where the service feature information includes at least one of the following information: RDMA connection type, RDMA operation type, RDMA transmission type, RDMA protocol type, service latency requirement, service bandwidth requirement, service priority requirement, or transmission rate; determine a first QoS requirement in the mapping relationship that matches the service feature information; the processing unit 1310 is further used to determine, based on the mapping relationship, a first QP in the at least one QP that corresponds to the first QoS requirement; and use the first QP to transmit a data packet that matches the service feature information.

[0339] In one possible implementation method, the transceiver unit 1320 is further used to receive a notification message from the control plane network element, where the notification message includes information about a first QP in the at least one QP, and the notification message is used to indicate the use of the first QP to transmit data packets of the first service; the processing unit 1310 is further used to transmit data packets of the first service using the first QP according to the notification message.

[0340] In a possible implementation method, the transceiver unit 1320 is further configured to send feedback information to the control plane network element, where the feedback information is used to indicate a transmission status of the RDMA link corresponding to the first QP.

[0341] When the communication device 1300 is used to implement the function of the control plane network element in the above method embodiment, the processing unit 1310 is used to control the transceiver unit 1320 to send first indication information to the first device, where the first indication information includes QP information of the first device, and the first indication information is used to indicate that a QP is established based on the QP information of the first device, and the QP information of the first device includes at least one of the following information: the number of QPs to be established, the memory area corresponding to the QP to be established, the access rights of the QP to be established, the type of the QP to be established, the maximum transmission unit of the QP to be established, and the transmission report of the QP to be established. The method further comprises: sending second indication information to the second device, the second indication information including QP information of the second device, the second indication information being used to indicate that a QP is established based on the QP information of the second device, the QP information of the second device including at least one of the following information: the number of QPs to be established, the memory area corresponding to the QP to be established, the access rights of the QP to be established, the type of the QP to be established, the maximum transmission unit of the QP to be established, the message length of the transmission of the QP to be established, or the mapping relationship between the QP to be established and the QoS requirement.

[0342] In one possible implementation method, the processing unit 1310 is further used to determine the QP information of the first device based on the information of the first device; wherein the information of the first device includes at least one of the hardware information of the first device, the memory information of the first device, or the RDMA information of the first device, and the RDMA information of the first device includes the RDMA protocol supported by the first device and the bearer network type corresponding to the supported RDMA protocol.

[0343] In one possible implementation method, the processing unit 1310 is further used to determine the QP information of the second device based on the information of the second device; wherein the information of the second device includes at least one of the hardware information of the second device, the memory information of the second device, or the RDMA information of the second device, and the RDMA information of the second device includes the RDMA protocol supported by the second device and the bearer network type corresponding to the supported RDMA protocol.

[0344] In a possible implementation method, the processing unit 1310 is further configured to control the transceiver unit 1320 to receive QP establishment information of the first device from the first device, where the QP establishment information of the first device is used to indicate information of the QP actually established by the first device.

[0345] In a possible implementation method, the processing unit 1310 is further configured to update the QP information of the first device according to the QP establishment information of the first device.

[0346] In a possible implementation method, the processing unit 1310 is further configured to control the transceiver unit 1320 to receive QP establishment information of the second device from the second device, where the QP establishment information of the second device is used to indicate information of the QP actually established by the second device.

[0347] In a possible implementation method, the processing unit 1310 is further configured to update the QP information of the second device according to the QP establishment information of the second device.

[0348] In one possible implementation method, the processing unit 1310 is further used to control the transceiver unit 1320 to receive feedback information from the first device or the second device, where the feedback information is used to indicate the transmission status of the RDMA link corresponding to the QP used when the first device and the second device perform data transmission; and based on the feedback information, control the transceiver unit 1320 to send an update indication to the first device or the second device, where the update indication is used to indicate an update of the used QP.

[0349] When the communication device 1300 is used to implement the function of the first device in the above method embodiment, the transceiver unit 1320 is used to receive indication information from the control plane network element, where the indication information includes at least one QoS requirement, and the indication information is used to indicate the establishment of a QP that meets the at least one QoS requirement; the processing unit 1310 is used to establish at least one QP that meets the at least one QoS requirement based on the indication information, and establish a mapping relationship between the at least one QoS requirement and the at least one QP.

[0350] In a possible implementation method, the indication information is further used to instruct establishment of an RDMA link between the first device and the second device; and the processing unit 1310 is further used to establish the RDMA link between the first device and the second device according to the at least one QP.

[0351] In one possible implementation method, the QoS requirement includes at least one of the following information: RDMA connection type, RDMA operation type, RDMA transmission type, RDMA protocol type, service latency requirement, service bandwidth requirement, service priority requirement or transmission rate.

[0352] In one possible implementation method, the transceiver unit 1320 is also used to send the QP establishment information of the first device to the control plane network element, where the QP establishment information of the first device is used to indicate the information of the QP actually established by the first device, and the information of the QP actually established by the first device includes the mapping relationship.

[0353] In one possible implementation method, the transceiver unit 1320 is further used to receive a PDR from the control plane network element, where the PDR includes service feature information of the first service, and the service feature information includes at least one of the following information: RDMA connection type, RDMA operation type, RDMA transmission type, RDMA protocol type, service latency requirement, service bandwidth requirement, service priority requirement, or transmission rate; the processing unit 1310 is further used to determine a first QoS requirement in the mapping relationship that matches the service feature information; determine a first QP corresponding to the first QoS requirement in the at least one QP based on the mapping relationship; and use the first QP to transmit a data packet that matches the service feature information.

[0354] In one possible implementation method, the transceiver unit 1320 is further used to receive a notification message from the control plane network element, where the notification message includes information about a first QP in the at least one QP, and the notification message is used to indicate the use of the first QP to transmit data packets of the first service; the processing unit 1310 is further used to transmit data packets of the first service using the first QP according to the notification message.

[0355] In a possible implementation method, the transceiver unit 1320 is further configured to send feedback information to the control plane network element, where the feedback information is used to indicate a transmission status of the RDMA link corresponding to the first QP.

[0356] When the communication device 1300 is used to implement the function of the control plane network element in the above method embodiment, the processing unit 1310 is used to control the transceiver unit 1320 to send first indication information to the first device, where the first indication information includes at least one QoS requirement, and the first indication information is used to indicate the establishment of a QP that meets the at least one QoS requirement; and send second indication information to the second device, where the second indication information includes the at least one QoS requirement, and the second indication information is used to indicate the establishment of a QP that meets the at least one QoS requirement.

[0357] In one possible implementation method, the processing unit 1310 is used to control the transceiver unit 1320 to send the first indication information to the first device, specifically including: controlling the transceiver unit 1320 to send the first indication information to the first device according to the information of the first device; wherein the information of the first device includes at least one of the hardware information of the first device, the memory information of the first device, or the RDMA information of the first device, and the RDMA information of the first device includes the RDMA protocol supported by the first device and the bearer network type corresponding to the supported RDMA protocol.

[0358] In one possible implementation method, the processing unit 1310 is used to control the transceiver unit 1320 to send the second indication information to the second device, specifically including: controlling the transceiver unit 1320 to send the second indication information to the second device according to the information of the second device; wherein the information of the second device includes at least one of the hardware information of the second device, the memory information of the second device, or the RDMA information of the second device, and the RDMA information of the second device includes the RDMA protocol supported by the second device and the bearer network type corresponding to the supported RDMA protocol.

[0359] In one possible implementation method, the processing unit 1310 is also used to control the transceiver unit 1320 to receive QP establishment information of the first device from the first device, where the QP establishment information of the first device is used to indicate information about the QP actually established by the first device, and the information about the QP actually established by the first device includes a mapping relationship between the at least one QoS requirement and the at least one QP established by the first device.

[0360] In a possible implementation method, the processing unit 1310 is further configured to update the QP information of the first device according to the QP establishment information of the first device.

[0361] In one possible implementation method, the processing unit 1310 is also used to control the transceiver unit 1320 to receive QP establishment information of the second device from the second device, where the QP establishment information of the second device is used to indicate information about the QP actually established by the second device, and the information about the QP actually established by the second device includes a mapping relationship between the at least one QoS requirement and the at least one QP established by the second device.

[0362] In a possible implementation method, the processing unit 1310 is further configured to update the QP information of the second device according to the QP establishment information of the second device.

[0363] In one possible implementation method, the processing unit 1310 is further used to control the transceiver unit 1320 to receive feedback information from the first device or the second device, where the feedback information is used to indicate the transmission status of the RDMA link corresponding to the QP used when the first device and the second device perform data transmission; and based on the feedback information, control the transceiver unit 1320 to send an update indication to the first device or the second device, where the update indication is used to indicate an update of the used QP.

[0364] A more detailed description of the processing unit 1310 and the transceiver unit 1320 can be directly obtained by referring to the relevant description in the above method embodiment, and will not be repeated here.

[0365] The communication device 1400 shown in Figure 14 includes a processor 1410 and an interface circuit 1420. The processor 1410 and the interface circuit 1420 are coupled to each other. It is understood that the interface circuit 1420 can be a transceiver or an input / output interface. Optionally, the communication device 1400 may also include a memory 1430 for storing instructions executed by the processor 1410, or storing input data required by the processor 1410 to execute instructions, or storing data generated after the processor 1410 executes instructions.

[0366] When the communication device 1400 is used to implement the above method embodiment, the processor 1410 is used to implement the functions of the above processing unit 1310 , and the interface circuit 1420 is used to implement the functions of the above transceiver unit 1320 .

[0367] It is understood that the processor in the embodiments of the present application may be a CPU, or may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. The general-purpose processor may be a microprocessor or any conventional processor.

[0368] The method steps in the embodiments of the present application can be implemented by hardware or by a processor executing software instructions. The software instructions can be composed of corresponding software modules, which can be stored in random access memory, flash memory, read-only memory, programmable read-only memory, erasable programmable read-only memory, electrically erasable programmable read-only memory, registers, hard disk, mobile hard disk, CD-ROM or any other form of storage medium well known in the art. An exemplary storage medium is coupled to the processor so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be an integral part of the processor. The processor and storage medium can be located in an ASIC. In addition, the ASIC can be located in an access network device or a terminal device. Of course, the processor and storage medium can also exist in the access network device or the terminal device as discrete components.

[0369] In the above embodiments, all or part of the embodiments may be implemented using software, hardware, firmware, or any combination thereof. When implemented using software, all or part of the embodiments may be implemented in the form of a computer program product. The computer program product includes one or more computer programs or instructions. When the computer program or instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present application are performed in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, an access network device, a terminal device, or other programmable device. The computer program or instructions may be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another. For example, the computer program or instructions may be transferred from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless means. The computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server or data center that integrates one or more available media. The available medium may be a magnetic medium, such as a floppy disk, hard disk, or magnetic tape; an optical medium, such as a digital video disk; or a semiconductor medium, such as a solid-state drive. The computer-readable storage medium may be a volatile or nonvolatile storage medium, or may include both volatile and nonvolatile types of storage media.

[0370] In the various embodiments of the present application, unless otherwise specified or there is a logical conflict, the terms and / or descriptions between different embodiments are consistent and can be referenced by each other. The technical features in different embodiments can be combined to form new embodiments according to their inherent logical relationships.

[0371] In this application, "at least one" means one or more, and "more" means two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone, where A and B can be singular or plural. In the text description of this application, the character " / " generally indicates that the previous and next related objects are in an "or" relationship; in the formulas of this application, the character " / " indicates that the previous and next related objects are in a "division" relationship.

[0372] It is understood that the various numbers used in the embodiments of this application are merely for ease of description and are not intended to limit the scope of the embodiments of this application. The order of the sequence numbers of the above-mentioned processes does not necessarily imply a specific order of execution; the order of execution of the processes should be determined by their functions and inherent logic.

Claims

1. A communication method, characterized in that, Applied to a first device or a module of the first device, the method includes: Receiving indication information from a control plane network element, where the indication information includes queue pair (QP) information of the first device, and the indication information is used to indicate establishing a QP based on the QP information of the first device. The QP information of the first device includes at least one of the following information: the number of QPs to be established, the memory area corresponding to the QPs to be established, the access permission of the QPs to be established, the type of the QPs to be established, the maximum transmission unit of the QPs to be established, the packet length of the transmission of the QPs to be established, or the mapping relationship between the QPs to be established and QoS requirements; Establishing at least one QP according to the indication information.

2. The method according to claim 1, wherein The indication information is further used to indicate establishing a Remote Direct Memory Access (RDMA) link between the first device and a second device; and establishing an RDMA link between the first device and the second device according to the at least one QP.

3. The method according to claim 1 or 2, characterized in that, The QoS requirements include at least one of the following information: RDMA connection type, RDMA operation type, RDMA transmission type, RDMA protocol type, service latency requirement, service bandwidth requirement, service priority requirement, or transmission rate.

4. The method according to any one of claims 1 to 3, characterized in that, The method further includes: Sending the QP establishment information of the first device to the control plane network element, where the QP establishment information of the first device is used to indicate the information of the QPs actually established by the first device.

5. The method according to any one of claims 1 to 4, characterized in that The method further includes: Receiving a Packet Detection Rule (PDR) from the control plane network element, where the PDR includes service characteristic information of a first service, and the service characteristic information includes at least one of the following information: RDMA connection type, RDMA operation type, RDMA transmission type, RDMA protocol type, service latency requirement, service bandwidth requirement, service priority requirement, or transmission rate; Determining a first QoS requirement in the mapping relationship that matches the service characteristic information; Determining a first QP in the at least one QPs that corresponds to the first QoS requirement according to the mapping relationship; Using the first QP to transmit data packets that match the service characteristic information.

6. The method according to any one of claims 1 to 4, characterized in that The method further includes: Receiving a notification message from the control plane network element, where the notification message includes information of a first QP in the at least one QPs, and the notification message is used to indicate using the first QP to transmit data packets of the first service; Using the first QP to transmit data packets of the first service according to the notification message.

7. The method according to claim 5 or 6, characterized in that, The method further includes: Sending feedback information to the control plane network element, where the feedback information is used to indicate the transmission status of the RDMA link corresponding to the first QP.

8. A communication method, characterized in that, Applied to a control plane network element or a module of the control plane network element, the method includes: Send first indication information to a first device, where the first indication information includes QP information of the first device, and the first indication information is used to indicate establishing a QP based on the QP information of the first device. The QP information of the first device includes at least one of the following information: the number of queue pairs QPs to be established, the memory area corresponding to the QP to be established, the access permission of the QP to be established, the type of the QP to be established, the maximum transmission unit of the QP to be established, the message length of the transmission of the QP to be established, or the mapping relationship between the QP to be established and QoS requirements; Send second indication information to the second device, where the second indication information includes QP information of the second device, and the second indication information is used to indicate establishing a QP based on the QP information of the second device. The QP information of the second device includes at least one of the following information: the number of queue pairs QPs to be established, the memory area corresponding to the QP to be established, the access permission of the QP to be established, the type of the QP to be established, the maximum transmission unit of the QP to be established, the message length of the transmission of the QP to be established, or the mapping relationship between the QP to be established and QoS requirements.

9. The method according to claim 8, wherein The method further includes: Determine the QP information of the first device according to the information of the first device; Wherein, the information of the first device includes at least one of the hardware information of the first device, the memory information of the first device, or the RDMA information of the first device. The RDMA information of the first device includes the RDMA protocol supported by the first device and the bearer network type corresponding to the supported RDMA protocol.

10. The method according to claim 8 or 9, characterized in that The method further includes: Determine the QP information of the second device according to the information of the second device; Wherein, the information of the second device includes at least one of the hardware information of the second device, the memory information of the second device, or the RDMA information of the second device. The RDMA information of the second device includes the RDMA protocol supported by the second device and the bearer network type corresponding to the supported RDMA protocol.

11. The method according to any one of claims 8 to 10, characterized in that, The method further includes: Receive the QP establishment information of the first device from the first device, where the QP establishment information of the first device is used to indicate the information of the QP actually established by the first device.

12. The method according to claim 11, wherein The method further includes: Update the QP information of the first device according to the QP establishment information of the first device.

13. The method according to any one of claims 8 to 12, characterized in that, The method further includes: Receive the QP establishment information of the second device from the second device, where the QP establishment information of the second device is used to indicate the information of the QP actually established by the second device.

14. The method according to claim 13, wherein The method further includes: Update the QP information of the second device according to the QP establishment information of the second device.

15. The method according to any one of claims 8 to 14, characterized in that, The method further includes: Receive feedback information from the first device or the second device, where the feedback information is used to indicate the transmission situation of the RDMA link corresponding to the QP used when the first device and the second device perform data transmission; Send an update indication to the first device or the second device according to the feedback information, where the update indication is used to indicate an update to the QP in use.

16. A communication method, characterized in that, Applied to a first device or a module of a first device, the method includes: Receive indication information from a control plane network element, where the indication information includes at least one quality of service (QoS) requirement, and the indication information is used to indicate the establishment of a QP that meets the at least one QoS requirement. According to the indication information, establish at least one QP that meets the at least one QoS requirement, and establish a mapping relationship between the at least one QoS requirement and the at least one QP.

17. The method according to claim 16, wherein The indication information is further used to indicate the establishment of a remote direct memory access (RDMA) link between the first device and a second device; the method further includes: Establish an RDMA link between the first device and the second device according to the at least one QP.

18. The method according to claim 16 or 17, characterized in that, The QoS requirement includes at least one of the following information: RDMA connection type, RDMA operation type, RDMA transmission type, RDMA protocol type, service latency requirement, service bandwidth requirement, service priority requirement, or transmission rate.

19. The method according to any one of claims 16 to 18, characterized in that, The method further includes: Send the QP establishment information of the first device to the control plane network element, where the QP establishment information of the first device is used to indicate the information of the QP actually established by the first device, and the information of the QP actually established by the first device includes the mapping relationship.

20. The method according to any one of claims 16 to 19, characterized in that The method further includes: Receive a packet detection rule (PDR) from the control plane network element, where the PDR includes service characteristic information of a first service, and the service characteristic information includes at least one of the following information: RDMA connection type, RDMA operation type, RDMA transmission type, RDMA protocol type, service latency requirement, service bandwidth requirement, service priority requirement, or transmission rate. Determine a first QoS requirement in the mapping relationship that matches the service characteristic information. According to the mapping relationship, determine a first QP in the at least one QP that corresponds to the first QoS requirement. Use the first QP to transmit data packets that match the service characteristic information.

21. The method according to any one of claims 16 to 19, characterized in that, The method further includes: Receive a notification message from the control plane network element, where the notification message includes information of a first QP in the at least one QP, and the notification message is used to indicate using the first QP to transmit data packets of a first service. According to the notification message, use the first QP to transmit data packets of the first service.

22. The method according to claim 20 or 21, wherein The method further includes: Send feedback information to the control plane network element, where the feedback information is used to indicate the transmission status of the RDMA link corresponding to the first QP.

23. A communication device, characterized in that, Includes a module for performing the method according to any one of claims 1 to 7, or performing the method according to any one of claims 8 to 15, or performing the method according to any one of claims 16 to 22.

24. A communication device, characterized in that, Comprising a processor and an interface circuit, the processor is used to communicate with other devices through the interface circuit and execute the method according to any one of claims 1 to 7, or execute the method according to any one of claims 8 to 15, or execute the method according to any one of claims 16 to 22.

25. A computer program product, characterized in that, The computer program product includes instructions, when the instructions run on a processor, causing the processor to execute the method according to any one of claims 1 to 7, or execute the method according to any one of claims 8 to 15, or execute the method according to any one of claims 16 to 22.

26. A computer-readable storage medium, characterized in that, The storage medium stores a computer program or instructions, when the computer program or instructions are executed by a communication device, implementing the method according to any one of claims 1 to 7, or implementing the method according to any one of claims 8 to 15, or implementing the method according to any one of claims 16 to 22.

27. A communication system, characterized in that, Comprising: A control plane network element, configured to send first indication information to a first device, the first indication information including QP information of the first device, the first indication information being used to indicate establishing a QP based on the QP information of the first device, the QP information of the first device including at least one of the following information: the number of queue pairs QPs to be established, the memory area corresponding to the QP to be established, the access permission of the QP to be established, the type of the QP to be established, the maximum transmission unit of the QP to be established, the message length of the transmission of the QP to be established, or the mapping relationship between the QP to be established and QoS requirements; The first device is configured to receive the first indication information; and establish at least one QP according to the first indication information.

28. The system according to claim 27, wherein, The communication system further includes a second device; The control plane network element is further configured to send second indication information to the second device, the second indication information including QP information of the second device, the second indication information being used to indicate establishing a QP based on the QP information of the second device, the QP information of the second device including at least one of the following information: the number of queue pairs QPs to be established, the memory area corresponding to the QP to be established, the access permission of the QP to be established, the type of the QP to be established, the maximum transmission unit of the QP to be established, the message length of the transmission of the QP to be established, or the mapping relationship between the QP to be established and QoS requirements; The second device is configured to receive the second indication information; and establish at least one QP according to the second indication information.

29. The system according to claim 28, wherein The first indication information is further used to indicate establishing an RDMA link between the first device and the second device, and the second indication information is further used to indicate establishing an RDMA link between the second device and the first device; The first device is further configured to establish at least one RDMA link with the second device according to at least one QP established by the first device; The second device is further configured to establish at least one RDMA link with the first device according to at least one QP established by the second device.

30. A communication system, characterized in that, Comprising: A control plane network element, configured to send first indication information to a first device, where the first indication information includes at least one Quality of Service (QoS) requirement, and the first indication information is used to indicate the establishment of a queue pair (QP) that meets the at least one QoS requirement. The first device is configured to receive the first indication information; according to the first indication information, establish at least one QP that meets the at least one QoS requirement, and establish a mapping relationship between the at least one QoS requirement and the at least one QP.

31. The system according to claim 30, wherein The communication system further includes a second device. The control plane network element is further configured to send second indication information to the second device, where the second indication information includes the at least one QoS requirement, and the second indication information is used to indicate the establishment of a QP that meets the at least one QoS requirement. The second device is configured to receive the second indication information. According to the second indication information, establish at least one QP that meets the at least one QoS requirement, and establish a mapping relationship between the at least one QoS requirement and the at least one QP.

32. The system according to claim 31, wherein The first indication information is further used to indicate the establishment of an RDMA link between the first device and the second device, and the second indication information is further used to indicate the establishment of an RDMA link between the second device and the first device. The first device is further configured to establish at least one RDMA link with the second device according to the at least one QP established by the first device. The second device is further configured to establish at least one RDMA link with the first device according to the at least one QP established by the second device.

Images