Quality of Service (QoS) measurement method, device and computing power node

Abstract

An embodiment of the present invention provides a method, device and computing power node for measuring quality of service (QoS). The method includes: performing QoS measurement based on computing power resources, or performing QoS measurement based on computing power resources and network resources. An embodiment of the present invention is directed to a computing power-aware network or a new network architecture of a computing power network, and uses computing power resources as a measurement indicator of QoS, and utilizes it in the corresponding network architecture to reserve or allocate more computing power resources for high computing power priority services when computing power resources are limited.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to a method, device and computing node for measuring quality of service (QoS). Background Art

[0002] Traditional networks only provide data communication channels, based on connections and subject to fixed network addressing mechanisms, which often fail to meet higher and more demanding quality of experience (QoE) requirements. With the vigorous development of cloud computing and edge computing, computing resources in the network are integrated into every corner of the network, so that each network node can become a resource provider. User requests can be met by calling the optimal node resources, no longer limited to a specific node, avoiding the waste of connection and network scheduling resources. This requires the ability of interactive scheduling between computing resource nodes, or the ability of dynamic routing of computing tasks, which can not only complete traditional high-load, long-connection tasks, but also handle lightweight, short-lifecycle computing tasks in a timely manner.

[0003] In today's networks, the network status and the resource information of the data center are not aware of each other. The network cannot perceive the service and resource deployment information within the data center. When a user requests a service, the network cannot send the service request to the best service node, which affects the user experience and reduces the efficiency of computing resources in the network.

[0004] Therefore, a new network architecture of computing power-aware network is proposed. It establishes a unified resource and service view for the computing power resources of the entire network, realizes mutual perception of network status and data center services and resource status, and dispatches user requests to the best data center for computing to ensure user experience.

[0005] The QoS (Quality of Service) measurement indicators of traditional networks mainly include bandwidth / throughput, latency, jitter, and packet loss rate. The traditional QoS measurement system is no longer applicable to the new computing power-aware network. Summary of the invention

[0006] The purpose of the embodiments of the present invention is to provide a method, device and computing power node for measuring quality of service (QoS) to solve the problem that the traditional QoS measurement system cannot be applied to the computing power perception system.

[0007] In order to solve the above problems, an embodiment of the present invention provides a quality of service (QoS) measurement method, which is applied to a computing power node, including:

[0008] QoS measurement is performed based on computing resources, or QoS measurement is performed based on computing resources and network resources.

[0009] Wherein, the method further comprises:

[0010] A tunnel corresponding to the service is established according to the computing resource requirements of at least one service; wherein at least one computing node in the tunnel reserves computing resources for the service.

[0011] Wherein, in the case where the computing power node is the head node of the tunnel, establishing a transmission tunnel corresponding to the service according to the computing power resource demand of at least one service includes:

[0012] Determining a computing power priority corresponding to the computing power resource demand according to the computing power resource demand of at least one business;

[0013] Deliver path messages to each computing node in the tunnel in sequence; the path messages carry: computing resource requirements of each service, computing priority corresponding to each computing resource requirement, and computing resource related information of the computing node that sends the path message;

[0014] Receive a reservation message transmitted in the opposite direction of the transmission direction of the path message; the reservation message carries information on computing resources that need to be reserved for each service in at least one computing node; wherein the path message and the reservation message are transmitted in a hop-by-hop manner.

[0015] Wherein, in the case where the computing power node is the tail node of the tunnel, establishing a transmission tunnel corresponding to the service according to the computing power resource demand of at least one service includes:

[0016] Receive a path message transmitted by hop-by-hop transmission from the head node of the tunnel, wherein the path message carries: computing power resource requirements of each service, computing power priority corresponding to each computing power resource requirement, and computing power resource related information of the computing power node sending the path message;

[0017] According to the path message, at least one computing node in the tunnel is selected to reserve computing resources for the corresponding service, and a reservation message is generated; the reservation message carries computing resource information that needs to be reserved for each service in at least one computing node;

[0018] The reservation message is transmitted in a hop-by-hop transmission manner in a direction opposite to the transmission direction of the path message.

[0019] Wherein, the method further comprises:

[0020] According to the path message and the current resource situation of the computing power node, corresponding computing power resources are reserved for the business.

[0021] Wherein, the method further comprises:

[0022] When the computing resources between the computing nodes of the tunnel change or are updated, the computing nodes dynamically interact with each other through the internal gateway protocol IGP.

[0023] The interactive message of the IGP includes: an optional sub-TLV, and the optional sub-TLV is used to announce and collect the reservable computing power resources corresponding to each computing power priority on the tunnel.

[0024] The IGP interaction message carries at least one of the following information:

[0025] Reserved computing resources information;

[0026] The maximum reservable computing resource information;

[0027] The computing power priority corresponding to the computing power resource information.

[0028] Wherein, the reservation message is a Resource Reservation Protocol RSVP message;

[0029] The RSVP message also includes at least one of the following:

[0030] The maximum computing power resource information of the computing power node;

[0031] The maximum reservable computing power resource information of the computing power node;

[0032] The currently available computing power resource information for each computing power priority on the computing power node;

[0033] The computing power priority of the computing power resources on this computing power node.

[0034] Wherein, the method further comprises:

[0035] Monitor the computing resource usage of the business on each computing node. When the computing resource usage meets the preset adjustment threshold, re-establish the tunnel corresponding to the business based on the computing resource requirements of at least one business.

[0036] The embodiment of the present invention further provides a QoS measurement device, which is applied to a computing power node, including:

[0037] The measurement module is used to perform QoS measurement based on computing resources, or to perform QoS measurement based on computing resources and network resources.

[0038] An embodiment of the present invention further provides a computing power node, including a processor and a transceiver, wherein the transceiver receives and sends data under the control of the processor, and the processor is used to perform the following operations:

[0039] QoS measurement is performed based on computing resources, or QoS measurement is performed based on computing resources and network resources.

[0040] The processor is further configured to perform the following operations:

[0041] A tunnel corresponding to the service is established according to the computing resource requirements of at least one service; wherein at least one computing node in the tunnel reserves computing resources for the service.

[0042] Wherein, when the computing power node is the head node of the tunnel, the processor is further configured to perform the following operations:

[0043] Determining a computing power priority corresponding to the computing power resource demand according to the computing power resource demand of at least one business;

[0044] Deliver path messages to each computing node in the tunnel in sequence; the path messages carry: computing resource requirements of each service, computing priority corresponding to each computing resource requirement, and computing resource related information of the computing node that sends the path message;

[0045] Receive a reservation message transmitted in the opposite direction of the transmission direction of the path message; the reservation message carries information on computing resources that need to be reserved for each service in at least one computing node; wherein the path message and the reservation message are transmitted in a hop-by-hop manner.

[0046] Wherein, when the computing power node is the tail node of the tunnel, the processor is further configured to perform the following operations:

[0047] Receive a path message transmitted by hop-by-hop transmission from the head node of the tunnel, wherein the path message carries: computing power resource requirements of each service, computing power priority corresponding to each computing power resource requirement, and computing power resource related information of the computing power node sending the path message;

[0048] According to the path message, at least one computing node in the tunnel is selected to reserve computing resources for the corresponding service, and a reservation message is generated; the reservation message carries computing resource information that needs to be reserved for each service in at least one computing node;

[0049] The reservation message is transmitted in a hop-by-hop transmission manner in a direction opposite to the transmission direction of the path message.

[0050] The processor is further configured to perform the following operations:

[0051] According to the path message and the current resource situation of the computing power node, corresponding computing power resources are reserved for the business.

[0052] The processor is further configured to perform the following operations:

[0053] When the computing resources between the computing nodes of the tunnel change or are updated, the computing nodes dynamically interact with each other through the internal gateway protocol IGP.

[0054] The interactive message of the IGP includes: an optional sub-TLV, and the optional sub-TLV is used to announce and collect the reservable computing power resources corresponding to each computing power priority on the tunnel.

[0055] The IGP interaction message carries at least one of the following information:

[0056] Reserved computing resources information;

[0057] The maximum reservable computing resource information;

[0058] The computing power priority corresponding to the computing power resource information.

[0059] Wherein, the reservation message is a Resource Reservation Protocol RSVP message;

[0060] The RSVP message also includes at least one of the following:

[0061] The maximum computing power resource information of the computing power node;

[0062] The maximum reservable computing power resource information of the computing power node;

[0063] The currently available computing power resource information for each computing power priority on the computing power node;

[0064] The computing power priority of the computing power resources on this computing power node.

[0065] The processor is further configured to perform the following operations:

[0066] Monitor the computing resource usage of the business on each computing node. When the computing resource usage meets the preset adjustment threshold, re-establish the tunnel corresponding to the business based on the computing resource requirements of at least one business.

[0067] An embodiment of the present invention also provides a computing power node, including a memory, a processor, and a program stored in the memory and executable on the processor, wherein the processor implements the quality of service QoS measurement method as described above when executing the program.

[0068] An embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the steps in the quality of service QoS measurement method described above are implemented.

[0069] The above technical solution of the present invention has at least the following beneficial effects:

[0070] In the quality of service QoS measurement method, device and computing power node of the embodiments of the present invention, computing power resources are used as QoS measurement indicators for computing power-aware networks or new network architectures of computing power networks, and are utilized in the corresponding network architecture to reserve or allocate more computing power resources for high computing power priority services when computing power resources are limited. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 A schematic diagram showing the steps of a method for measuring quality of service (QoS) provided in an embodiment of the present invention;

[0072] Figure 2 An example of computing resource allocation corresponding to computing priority in the quality of service QoS measurement method provided by an embodiment of the present invention is shown;

[0073] Figure 3 A schematic diagram showing the format of an IGP message in a quality of service (QoS) measurement method provided in an embodiment of the present invention;

[0074] Figure 4 A schematic diagram showing a flow chart of Example 1 provided in an embodiment of the present invention;

[0075] Figure 5 A schematic diagram showing the structure of a device for measuring quality of service (QoS) provided in an embodiment of the present invention;

[0076] Figure 6 A schematic diagram showing the structure of a computing power node provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0077] In order to make the technical problems, technical solutions and advantages to be solved by the present invention more clear, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0078] like Figure 1 As shown, an embodiment of the present invention provides a method for measuring quality of service (QoS), which is applied to a computing power node and includes:

[0079] Step 101, perform QoS measurement based on computing resources, or perform QoS measurement based on computing resources and network resources.

[0080] Among them, the computing power resources include at least one of the following: central processing unit CPU, graphics processing unit GPU, storage information, memory information and cache information; the network resources include at least one of the following: delay information, jitter information, bandwidth information, throughput information, packet loss rate and reliability.

[0081] Furthermore, at least one embodiment of the present invention also provides a resource reservation method for QoS measurement based on computing resources (or computing resources and network resources), and the protocol technology used is MPLS DS-TE (Multi-Protocol Label Switching Differentiated Services Flow Engineering) technology. That is, in an embodiment of the present invention, the method also includes:

[0082] A tunnel corresponding to the service is established according to the computing resource requirements of at least one service; wherein at least one computing node in the tunnel reserves computing resources for the service.

[0083] It should be noted that in order to use MPLS DS-TE to reserve different computing power for different service types, the embodiment of the present invention introduces the concept of computing power priority (Compute Class, CC). The computing power set corresponding to a group of QoS flows with specified computing power requirements is called a CC (for example, a computing power of 50 to 80 Tflops is required). Computing power resources can be allocated separately for each CC, such as Figure 2 shown.

[0084] For example, according to the required level of computing power, the number of computing power priority CCs is defined as 3, as shown in the following table:

[0085] CC Hash power level Task Type CC0-High Computationally intensive For example, deep learning tasks CC1-Medium Computing Medium Such as image rendering CC2-Low Compute sparse type General computing business processing

[0086] At the same time, two computing power constraint models are proposed, namely, the computing power constraint model when the computing power node reserves computing power resources for the business:

[0087] 1. CC0, CC1, and CC2 do not preempt each other and only use the original reserved computing resources;

[0088] 2. CC0, CC1, and CC2 preempt each other, and high-priority resources can preempt low-priority computing resources. For example, CC0, as the highest priority, can preempt low-priority resources.

[0089] As an optional embodiment, when the computing power node is the head node of the tunnel, establishing a transmission tunnel corresponding to the service according to the computing power resource requirement of at least one service includes:

[0090] Determining a computing power priority corresponding to the computing power resource demand according to the computing power resource demand of at least one business;

[0091] Deliver path messages to each computing node in the tunnel in sequence; the path messages carry: computing resource requirements of each service, computing priority corresponding to each computing resource requirement, and computing resource related information of the computing node that sends the path message;

[0092] Receive a reservation message transmitted in the opposite direction of the transmission direction of the path message; the reservation message carries information on computing resources that need to be reserved for each service in at least one computing node; wherein the path message and the reservation message are transmitted in a hop-by-hop manner.

[0093] As another optional embodiment, when the computing power node is the tail node of the tunnel, establishing a transmission tunnel corresponding to the service according to the computing power resource demand of at least one service includes:

[0094] Receive a path message transmitted by hop-by-hop transmission from the head node of the tunnel, wherein the path message carries: computing power resource requirements of each service, computing power priority corresponding to each computing power resource requirement, and computing power resource related information of the computing power node sending the path message;

[0095] According to the path message, at least one computing node in the tunnel is selected to reserve computing resources for the corresponding service, and a reservation message is generated; the reservation message carries computing resource information that needs to be reserved for each service in at least one computing node;

[0096] The reservation message is transmitted in a hop-by-hop transmission manner in a direction opposite to the transmission direction of the path message.

[0097] Following the above example, the method further includes:

[0098] The computing power node that receives the reservation message reserves corresponding computing power resources for the business based on the path message and the current resource situation of the computing power node.

[0099] For example, each computing power node on the tunnel decides whether to accept the tunnel based on the current resource situation. If the tunnel is accepted, resources are reserved at the corresponding computing power priority; otherwise, no computing power resources are reserved.

[0100] It should be noted that when each computing power node reserves corresponding computing power resources for the business, it is necessary to comply with the above-mentioned computing power constraint model. It can only use the reserved computing power resources corresponding to the computing power priority, or high priority can preempt low priority computing power resources to reserve resources.

[0101] It should be noted that the tunnel can only be established if the computing nodes through which the path message and the reserved message pass meet the computing resource requirements of the business; otherwise, the tunnel establishment fails, and the computing business application needs to resend the computing resource requirements to re-trigger the establishment of the tunnel.

[0102] As an optional embodiment of the present invention, the method further includes:

[0103] When the computing resources between the computing nodes of the tunnel change or are updated, the computing nodes dynamically interact with each other through the internal gateway protocol IGP.

[0104] Among them, the interactive message of the IGP includes: optional sub-TLV (type length content), which is used to announce and collect the reservable computing resources corresponding to each computing priority on the tunnel. For example, the newly added optional sub-TLV type is 30, the length is 40Byte, and it carries the reserved computing resources, the maximum reservable resources, the corresponding computing level, etc. That is, the interactive message of the IGP carries at least one of the following information:

[0105] Reserved computing resources information;

[0106] The maximum reservable computing resource information;

[0107] The computing power priority corresponding to the computing power resource information.

[0108] For another example, in the sub-TLV that has been defined in the IGP ISIS protocol sub-TLV type, a new optional sub-TLV is introduced. The specific format is as follows Figure 3 As shown, the CC Model ID field is used to indicate the computing resource ID, the Reserved field is a reserved field, the CC0value field is used to carry the computing resource information corresponding to the computing priority CC0, the CC1 value field is used to carry the computing resource information corresponding to the computing priority CC1, and the CC2value field is used to carry the computing resource information corresponding to the computing priority CC2.

[0109] As another optional embodiment, the reservation message is a Resource Reservation Protocol RSVP message;

[0110] The RSVP message also includes at least one of the following:

[0111] The maximum computing power resource information of the computing power node;

[0112] The maximum reservable computing power resource information of the computing power node;

[0113] The currently available computing power resource information for each computing power priority on the computing power node;

[0114] The computing power priority of the computing power resources on this computing power node.

[0115] For example, ClassNum 30 is defined as the computing power priority, and the RSVP message carries specific computing power data indicators, such as the maximum computing power resources, the maximum reservable computing power resources, the current available resources of each priority, the computing power level, etc.

[0116] As yet another optional embodiment, the method further includes:

[0117] Monitor the computing resource usage of the business on each computing node. When the computing resource usage meets the preset adjustment threshold, re-establish the tunnel corresponding to the business based on the computing resource requirements of at least one business.

[0118] For example, if resource change value / current usage value*100%>threshold value, the step of re-establishing the tunnel is initiated.

[0119] In the embodiment of the present invention, a new concept of computing power priority is defined, and field extension is performed in protocols such as IGP and RSVP in MPLS DS-TE technology to realize information transmission on the control plane, thereby ensuring an end-to-end priority-based computing power resource reservation and preemption mechanism.

[0120] In order to more clearly describe the computing power resource reservation method provided in an embodiment of the present invention, an example is provided below for illustration.

[0121] Example 1: Figure 4 As shown, the methods for reserving computing resources include:

[0122] Step 1: Services 1, 2, and 3 of the computing service application carry their respective computing resource requirements and send them to access router R1 (i.e., computing node, generally the head node of the computing network). (Specifically, it can be carried through the TLV field in SRv6, or pre-configured on the R1 device by the administrator).

[0123] Step 2: R1 maps computing resources to specific CC level values ​​and saves the corresponding relationship;

[0124] Step 3: Configure the MPLS DS-TE tunnel. The RSVP protocol carries the CC level and computing resources required to be reserved for each service through the Path message, which is transmitted from R1 to R4 along the path. R4 aggregates all the information and performs intelligent calculations to select the best node or nodes in the path to reserve resources.

[0125] Step 4: The reservation message (Resv message) in the RSVP protocol is sent in the reverse direction of the data flow, from R4 to R1 in sequence. It carries the specific resources that each service needs to reserve at each node. Each node reserves resources for the three services (possibly three different CCs) at the same node or multiple nodes based on this information.

[0126] Step 5: The resource changes and updates between the nodes on the path are dynamically diffused through the IGP protocol.

[0127] In summary, in the face of a new type of computing power-aware network, the embodiments of the present invention add computing power resources as a QoS measurement system and utilize them in corresponding network technologies to reserve or allocate more computing power resources for high-priority services when computing power resources are limited, to meet the current users' diverse needs for cloud-network integration, and to ensure that different levels of business types can give priority to using more computing power resources.

[0128] like Figure 5 As shown, an embodiment of the present invention further provides a quality of service QoS measurement device, which is applied to a computing power node, including:

[0129] The measurement module 501 is used to perform QoS measurement based on computing resources, or to perform QoS measurement based on computing resources and network resources.

[0130] As an optional embodiment, the device further includes:

[0131] A module is established to establish a tunnel corresponding to a service according to the computing power resource requirements of at least one service; wherein at least one computing power node in the tunnel reserves computing power resources for the service.

[0132] As an optional embodiment, when the computing power node is the head node of the tunnel, the establishing module includes:

[0133] A first submodule is used to determine a computing power priority corresponding to the computing power resource demand according to the computing power resource demand of at least one business;

[0134] The second submodule is used to sequentially transmit path messages to each computing power node in the tunnel; the path message carries: the computing power resource requirements of each business, the computing power priority corresponding to each computing power resource requirement, and computing power resource related information of the computing power node sending the path message;

[0135] The third submodule is used to receive a reservation message transmitted in the opposite direction of the transmission direction of the path message; the reservation message carries the computing resource information that needs to be reserved for each service in at least one computing node; wherein the path message and the reservation message are transmitted in a hop-by-hop manner.

[0136] As an optional embodiment, when the computing power node is the tail node of the tunnel, the establishing module includes:

[0137] The fourth submodule is used to receive a path message transmitted by hop-by-hop transmission from the head node of the tunnel, wherein the path message carries: computing power resource requirements of each business, computing power priority corresponding to each computing power resource requirement, and computing power resource related information of the computing power node that sends the path message;

[0138] A fifth submodule is used to select at least one computing node in the tunnel to reserve computing resources for the corresponding business according to the path message, and generate a reservation message; the reservation message carries computing resource information that each business needs to reserve in at least one computing node;

[0139] The sixth submodule is used to transmit the reservation message in a hop-by-hop transmission manner in a direction opposite to the transmission direction of the path message.

[0140] As an optional embodiment, the device further includes:

[0141] The reservation module is used to reserve corresponding computing resources for the business according to the path message and the current resource situation of the computing node.

[0142] As an optional embodiment, the device further includes:

[0143] The interaction module is used to dynamically interact between the computing nodes through the internal gateway protocol IGP when the computing resources between the computing nodes of the tunnel change or are updated.

[0144] The first submodule is used for the interactive message of the IGP to include: an optional sub-TLV, and the optional sub-TLV is used to announce and collect the reservable computing power resources corresponding to each computing power priority on the tunnel.

[0145] As an optional embodiment, the IGP interaction message carries at least one of the following information:

[0146] Reserved computing resources information;

[0147] The maximum reservable computing resource information;

[0148] The computing power priority corresponding to the computing power resource information.

[0149] As an optional embodiment, the reservation message is a Resource Reservation Protocol RSVP message;

[0150] The RSVP message also includes at least one of the following:

[0151] The maximum computing power resource information of the computing power node;

[0152] The maximum reservable computing power resource information of the computing power node;

[0153] The currently available computing power resource information for each computing power priority on the computing power node;

[0154] The computing power priority of the computing power resources on this computing power node.

[0155] As an optional embodiment, the device further includes:

[0156] The trigger module is used to monitor the computing power resource usage of the business on each computing power node. When the computing power resource usage meets the pre-set adjustment threshold, the tunnel corresponding to the business is re-established according to the computing power resource requirements of at least one business.

[0157] In the face of a new type of computing power-aware network, the embodiments of the present invention add computing power resources as a QoS measurement system and utilize them in corresponding network technologies to reserve or allocate more computing power resources for high-priority services when computing power resources are limited, to meet the current users' diverse needs for cloud-network integration, and to ensure that different levels of business types can give priority to using more computing power resources.

[0158] It should be noted that the quality of service QoS measurement device provided in the embodiment of the present invention is a device capable of executing the above-mentioned quality of service QoS measurement method, so all embodiments of the above-mentioned quality of service QoS measurement method are applicable to the device and can achieve the same or similar beneficial effects.

[0159] like Figure 6 As shown, an embodiment of the present invention further provides a computing power node, including a processor 600 and a transceiver 610, wherein the transceiver 610 receives and sends data under the control of the processor 600, and the processor 600 is used to perform the following operations:

[0160] QoS measurement is performed based on computing resources, or QoS measurement is performed based on computing resources and network resources.

[0161] As an optional embodiment, the processor is further configured to perform the following operations:

[0162] A tunnel corresponding to the service is established according to the computing resource requirements of at least one service; wherein at least one computing node in the tunnel reserves computing resources for the service.

[0163] As an optional embodiment, when the computing power node is the head node of the tunnel, the processor is further configured to perform the following operations:

[0164] Determining a computing power priority corresponding to the computing power resource demand according to the computing power resource demand of at least one business;

[0165] Deliver path messages to each computing node in the tunnel in sequence; the path messages carry: computing resource requirements of each service, computing priority corresponding to each computing resource requirement, and computing resource related information of the computing node that sends the path message;

[0166] Receive a reservation message transmitted in the opposite direction of the transmission direction of the path message; the reservation message carries information on computing resources that need to be reserved for each service in at least one computing node; wherein the path message and the reservation message are transmitted in a hop-by-hop manner.

[0167] As an optional embodiment, when the computing power node is the tail node of the tunnel, the processor is further configured to perform the following operations:

[0168] Receive a path message transmitted by hop-by-hop transmission from the head node of the tunnel, wherein the path message carries: computing power resource requirements of each service, computing power priority corresponding to each computing power resource requirement, and computing power resource related information of the computing power node sending the path message;

[0169] According to the path message, at least one computing node in the tunnel is selected to reserve computing resources for the corresponding service, and a reservation message is generated; the reservation message carries computing resource information that needs to be reserved for each service in at least one computing node;

[0170] The reservation message is transmitted in a hop-by-hop transmission manner in a direction opposite to the transmission direction of the path message.

[0171] As an optional embodiment, the processor is further configured to perform the following operations:

[0172] According to the path message and the current resource situation of the computing power node, corresponding computing power resources are reserved for the business.

[0173] As an optional embodiment, the processor is further configured to perform the following operations:

[0174] When the computing resources between the computing nodes of the tunnel change or are updated, the computing nodes dynamically interact with each other through the internal gateway protocol IGP.

[0175] As an optional embodiment, the interaction message of the IGP includes: an optional sub-TLV, and the optional sub-TLV is used to announce and collect reservable computing power resources corresponding to each computing power priority on the tunnel.

[0176] As an optional embodiment, the IGP interaction message carries at least one of the following information:

[0177] Reserved computing resources information;

[0178] The maximum reservable computing resource information;

[0179] The computing power priority corresponding to the computing power resource information.

[0180] As an optional embodiment, the reservation message is a Resource Reservation Protocol RSVP message;

[0181] The RSVP message also includes at least one of the following:

[0182] The maximum computing power resource information of the computing power node;

[0183] The maximum reservable computing power resource information of the computing power node;

[0184] The currently available computing power resource information for each computing power priority on the computing power node;

[0185] The computing power priority of the computing power resources on this computing power node.

[0186] As an optional embodiment, the processor is further configured to perform the following operations:

[0187] Monitor the computing resource usage of the business on each computing node. When the computing resource usage meets the preset adjustment threshold, re-establish the tunnel corresponding to the business based on the computing resource requirements of at least one business.

[0188] In the face of a new type of computing power-aware network, the embodiments of the present invention add computing power resources as a QoS measurement system and utilize them in corresponding network technologies to reserve or allocate more computing power resources for high-priority services when computing power resources are limited, to meet the current users' diverse needs for cloud-network integration, and to ensure that different levels of business types can give priority to using more computing power resources.

[0189] It should be noted that the computing power node provided in the embodiment of the present invention is a computing power node capable of executing the above-mentioned quality of service QoS measurement method. All embodiments of the above-mentioned quality of service QoS measurement method are applicable to the computing power node and can achieve the same or similar beneficial effects.

[0190] An embodiment of the present invention also provides a computing power node, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, each process in the embodiment of the quality of service (QoS) measurement method as described above is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be described here.

[0191] The embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, each process in the above-mentioned quality of service QoS measurement method embodiment is implemented, and the same technical effect can be achieved. To avoid repetition, it is not repeated here. The computer-readable storage medium is, for example, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0192] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-readable storage media (including but not limited to disk storage and optical storage, etc.) containing computer-usable program code.

[0193] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A device that specifies functions in one or more processes and / or one or more blocks.

[0194] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable storage medium produce a paper product including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0195] These computer program instructions may also be loaded onto a computer or other programmable data processing device so that the computer or other programmable device executes a series of operating steps to produce a computer-implemented process, thereby providing instructions executed on the computer or other programmable device for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0196] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A quality of service (QoS) measurement method, applied to computing nodes, characterized in that: include: QoS measurement is performed based on computing resources, or QoS measurement is performed based on computing resources and network resources; Among them, the resources for QoS measurement based on computing resources or computing resources and network resources are reserved, including: According to the computing resource requirements of at least one business, a tunnel corresponding to the business is established; wherein at least one computing node in the tunnel reserves computing resources for the business; Wherein, in the case where the computing power node is the head node of the tunnel, establishing a transmission tunnel corresponding to the service according to the computing power resource demand of at least one service includes: Determining a computing power priority corresponding to the computing power resource demand according to the computing power resource demand of at least one business; Deliver path messages to each computing node in the tunnel in sequence; the path messages carry: computing resource requirements of each service, computing priority corresponding to each computing resource requirement, and computing resource related information of the computing node that sends the path message; Receiving a reservation message transmitted in the opposite direction of the transmission direction of the path message; the reservation message carries information about computing resources that each service needs to reserve in at least one computing node; wherein the path message and the reservation message are transmitted in a hop-by-hop transmission manner; Alternatively, when the computing power node is the tail node of the tunnel, establishing a transmission tunnel corresponding to the service according to the computing power resource requirement of at least one service includes: Receive a path message transmitted by hop-by-hop transmission from the head node of the tunnel, wherein the path message carries: computing power resource requirements of each service, computing power priority corresponding to each computing power resource requirement, and computing power resource related information of the computing power node sending the path message; According to the path message, at least one computing node in the tunnel is selected to reserve computing resources for the corresponding service, and a reservation message is generated; the reservation message carries computing resource information that needs to be reserved for each service in at least one computing node; The reservation message is transmitted in a hop-by-hop transmission manner in a direction opposite to the transmission direction of the path message.

2. The method according to claim 1, characterized in that The method further comprises: According to the path message and the current resource situation of the computing power node, corresponding computing power resources are reserved for the business.

3. The method according to claim 1, characterized in that The method further comprises: When the computing resources between the computing nodes of the tunnel change or are updated, the computing nodes dynamically interact with each other through the internal gateway protocol IGP.

4. The method according to claim 3, characterized in that: The interactive message of the IGP includes: an optional sub-TLV, where the optional sub-TLV is used to announce and collect reservable computing resources corresponding to each computing priority on the tunnel.

5. The method according to claim 4, characterized in that The IGP interaction message carries at least one of the following information: Reserved computing resources information; The maximum reservable computing power resource information; The computing power priority corresponding to the computing power resource information.

6. The method according to claim 1, characterized in that The reservation message is a Resource Reservation Protocol RSVP message; The RSVP message also includes at least one of the following: The maximum computing power resource information of the computing power node; The maximum reservable computing power resource information of the computing power node; The currently available computing power resource information for each computing power priority on the computing power node; The computing power priority of the computing power resources on this computing power node.

7. The method according to claim 1, characterized in that The method further comprises: Monitor the computing resource usage of the business on each computing node. When the computing resource usage meets the preset adjustment threshold, re-establish the tunnel corresponding to the business based on the computing resource requirements of at least one business.

8. A QoS measurement device, applied to a computing node, characterized in that: include: A measurement module, used to perform QoS measurement based on computing resources, or to perform QoS measurement based on computing resources and network resources; Wherein, the device further comprises: An establishment module is used to establish a tunnel corresponding to the service according to the computing power resource requirements of at least one service; wherein at least one computing power node in the tunnel reserves computing power resources for the service; Wherein, when the computing power node is the head node of the tunnel, the establishing module includes: A first submodule is used to determine a computing power priority corresponding to the computing power resource demand according to the computing power resource demand of at least one business; The second submodule is used to sequentially transmit path messages to each computing power node in the tunnel; the path message carries: the computing power resource requirements of each business, the computing power priority corresponding to each computing power resource requirement, and computing power resource related information of the computing power node sending the path message; The third submodule is used to receive a reservation message transmitted in the opposite direction of the transmission direction of the path message; the reservation message carries information about computing resources that each service needs to reserve in at least one computing node; wherein the path message and the reservation message are transmitted in a hop-by-hop transmission manner; Alternatively, when the computing power node is the tail node of the tunnel, the establishing module includes: The fourth submodule is used to receive a path message transmitted by hop-by-hop transmission from the head node of the tunnel, wherein the path message carries: computing power resource requirements of each business, computing power priority corresponding to each computing power resource requirement, and computing power resource related information of the computing power node that sends the path message; A fifth submodule is used to select at least one computing node in the tunnel to reserve computing resources for the corresponding business according to the path message, and generate a reservation message; the reservation message carries computing resource information that each business needs to reserve in at least one computing node; The sixth submodule is used to transmit the reservation message in a hop-by-hop transmission manner in a direction opposite to the transmission direction of the path message.

9. A computing node, comprising a processor and a transceiver, wherein the transceiver receives and sends data under the control of the processor, characterized in that: The processor is configured to perform the following operations: QoS measurement is performed based on computing resources, or QoS measurement is performed based on computing resources and network resources; Among them, the resources for QoS measurement based on computing resources or computing resources and network resources are reserved, including: According to the computing resource requirements of at least one business, a tunnel corresponding to the business is established; wherein at least one computing node in the tunnel reserves computing resources for the business; Wherein, when the computing power node is the head node of the tunnel, the processor is further configured to perform the following operations: Determining a computing power priority corresponding to the computing power resource demand according to the computing power resource demand of at least one business; Deliver path messages to each computing node in the tunnel in sequence; the path messages carry: computing resource requirements of each service, computing priority corresponding to each computing resource requirement, and computing resource related information of the computing node that sends the path message; Receiving a reservation message transmitted in the opposite direction of the transmission direction of the path message; the reservation message carries information about computing resources that each service needs to reserve in at least one computing node; wherein the path message and the reservation message are transmitted in a hop-by-hop transmission manner; Alternatively, when the computing power node is the tail node of the tunnel, the processor is further configured to perform the following operations: Receive a path message transmitted by hop-by-hop transmission from the head node of the tunnel, wherein the path message carries: computing power resource requirements of each service, computing power priority corresponding to each computing power resource requirement, and computing power resource related information of the computing power node sending the path message; According to the path message, at least one computing node in the tunnel is selected to reserve computing resources for the corresponding service, and a reservation message is generated; the reservation message carries computing resource information that needs to be reserved for each service in at least one computing node; The reservation message is transmitted in a hop-by-hop transmission manner in a direction opposite to the transmission direction of the path message.

10. The computing power node according to claim 9, characterized in that: The processor is further configured to perform the following operations: According to the path message and the current resource situation of the computing power node, corresponding computing power resources are reserved for the business.

11. The computing power node according to claim 9, characterized in that: The processor is further configured to perform the following operations: When the computing resources between the computing nodes of the tunnel change or are updated, the computing nodes dynamically interact with each other through the internal gateway protocol IGP.

12. The computing power node according to claim 11, characterized in that: The interactive message of the IGP includes: an optional sub-TLV, where the optional sub-TLV is used to announce and collect reservable computing resources corresponding to each computing priority on the tunnel.

13. The computing node according to claim 11, characterized in that: The IGP interaction message carries at least one of the following information: Reserved computing resources information; The maximum reservable computing power resource information; The computing power priority corresponding to the computing power resource information.

14. The computing power node according to claim 9, characterized in that: The reservation message is a Resource Reservation Protocol RSVP message; The RSVP message also includes at least one of the following: The maximum computing power resource information of the computing power node; The maximum reservable computing power resource information of the computing power node; The currently available computing power resource information for each computing power priority on the computing power node; The computing power priority of the computing power resources on this computing power node.

15. The computing power node according to claim 9, characterized in that: The processor is further configured to perform the following operations: Monitor the computing resource usage of the business on each computing node. When the computing resource usage meets the preset adjustment threshold, re-establish the tunnel corresponding to the business based on the computing resource requirements of at least one business.

16. A computing node, comprising a memory, a processor, and a program stored in the memory and executable on the processor; characterized in that: When the processor executes the program, the quality of service (QoS) measurement method according to any one of claims 1 to 7 is implemented.

17. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps in the quality of service QoS measurement method according to any one of claims 1 to 7 are implemented.

Images