Message sending method, target node, and electronic device
By reserved offset outbound time slots in the target node and converted into time slots in the scheduling cycle, the network congestion problem caused by the orchestration period length is solved, and the packet forwarding efficiency is improved.
Patent Information
- Application Number
- PCT/CN2024/072562
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2023-05-11
- Filing Date
- 2024-01-16
- Publication Date
- 2024-11-14
AI Technical Summary
When the orchestration period length is greater than the scheduling period length, the target node is prone to generate a large number of queues or cache requirements, resulting in network congestion and affecting the efficiency of message forwarding.
By reserved a first target outbound time slot for the target service in the target node, the number of offsets of the time slot relative to the time slot in the transmission in progress is less than the number of first time slots in the scheduling cycle, and upon receiving After forwarding the message, based on the preset conversion strategy, the first target outbound time slot is converted into the second target outbound time slot in the scheduling cycle, and the forwarded message is stored in the queue to be sent to the second target outbound time slot.
It effectively reduces the queue and cache requirements of the target nodes, reduces the possibility of network congestion, and improves the forwarding efficiency of packets.
Smart Images

Figure CN2024072562_14112024_PF_FP_ABST
Abstract
Description
Message sending method, target node and electronic device
[0001] Cross-references
[0002] This application claims priority to the Chinese patent application filed with the China Patent Office on May 11, 2023, with application number 202310534244.7 and application name “Message Sending Method, Target Node and Electronic Device”. The entire contents of the application are incorporated by reference into this application. Technical Field
[0003] The embodiments of the present application relate to the field of data communication technology, and in particular to a message sending method, a target node, and an electronic device. Background Art
[0004] The Internet Engineering Task Force (IETF) standard RFC8655 describes the architecture of deterministic networks and defines the QoS (Quality of Service) objectives for deterministic forwarding. To achieve these QoS objectives, the transmission paths for service flows in the network can be preselected to ensure routing stability that persists despite real-time changes in the network topology.
[0005] Typically, the deterministic path preselected for a service flow is a strictly explicit path calculated by a centralized controller. Resources are reserved at nodes along the path to meet the SLA (Service Level Agreement) requirements of the deterministic service. For example, taking two adjacent nodes along a service transmission path, the outbound time slot of the upstream node's egress port is mapped to the in-progress time slot of the downstream node's egress port. Based on this mapping, outbound time slot resources can be reserved for the service flow at the downstream node's egress port. Forwarded messages are then sent sequentially in the outbound time slot of the egress port of each node in the deterministic path.
[0006] However, since the orchestration cycle is business-oriented, the length of the orchestration cycle is usually very long. If the node implements a scheduling cycle of the same length on the forwarding plane, it will generate a large number of queues or cache requirements, even far exceeding the node's carrying capacity. At this time, network congestion is likely to occur, seriously affecting the forwarding efficiency of the message.
[0007] Summary of the Invention
[0008] The embodiments of the present application provide a message sending method, a target node, and an electronic device to at least solve the problem in the related art that, when the length of the orchestration cycle is longer than the length of the scheduling cycle, the target node generates a large number of queues or cache requirements, which easily causes network congestion and affects the forwarding efficiency of the message.
[0009] In order to solve the above technical problems, this application is implemented as follows:
[0010] In the first aspect, an embodiment of the present application provides a message sending method, including: a target node reserves a first target outgoing time slot for a target service within a scheduling cycle, wherein the number of time slot offsets of the first target outgoing time slot relative to the target node's sending in-progress time slot is less than the first number of time slots included in the scheduling cycle, and the target node is any node on the target path of the target service flow; after receiving a forwarding message of the target service flow, the target node converts the first target outgoing time slot into a second target outgoing time slot in the scheduling cycle based on a preset conversion strategy; the target node stores the forwarding message in a to-be-sent queue corresponding to the second target outgoing time slot for sending.
[0011] In one possible implementation, after receiving the forwarding message of the target business flow, the target node converts the first target outgoing time slot into the second target outgoing time slot in the scheduling period based on a preset conversion strategy, including: when the target node receives the forwarding message of the target business flow, obtaining the first target outgoing time slot corresponding to the forwarding message; based on the preset conversion strategy, converting the first target outgoing time slot into the second target outgoing time slot in the scheduling period.
[0012] In a possible implementation, obtaining the first target outgoing time slot corresponding to the forwarded message includes: determining, based on the time slot number carried by the forwarded message, the first target outgoing time slot reserved by the target node for the target service flow within the orchestration cycle, the time slot number corresponding to the first target outgoing time slot.
[0013] In a possible implementation, obtaining the first target outgoing time slot corresponding to the forwarded message includes: determining the outgoing time slot corresponding to the forwarded message based on a pre-stored message forwarding table entry; and determining the outgoing time slot as the first target outgoing time slot reserved by the target node for the target service flow within the orchestration cycle.
[0014] In one possible implementation, the preset conversion strategy includes: when the number of second time slots included in the scheduling period is an integer multiple of the number of first time slots included in the scheduling period, the first target outgoing time slot performs a modulo operation on the first number of time slots to obtain the second target outgoing time slot.
[0015] In one possible implementation, the target node reserves a first target outbound time slot for the target service within the scheduling cycle, including: obtaining time slot resource information of the target node within the scheduling cycle, and the first number of time slots included in the scheduling cycle of the target node; reserving a target time slot for the target service based on a predetermined residence delay budget value corresponding to the target node and the time slot resource information, and determining a residence delay evaluation value corresponding to the target node; when the difference between the residence delay budget value and the residence delay evaluation value is less than a preset threshold, and the number of time slot offsets of the target time slot relative to the sending-in-progress time slot of the target node is less than the first number of time slots, determining the target time slot as the first target outbound time slot reserved by the target node for the target service within the scheduling cycle; wherein the sending-in-progress time slot is used to indicate a time slot in a sending state on the egress port of the target node.
[0016] In one possible implementation, obtaining the time slot resource information of the target node within the scheduling period and the first number of time slots included in the scheduling period of the target node includes: notifying the time slot resource information of the target node within the scheduling period and the length information of the scheduling period through an extended network protocol; the length information is used to indicate the first number of time slots included in the scheduling period or the length of the scheduling period.
[0017] In the second aspect, an embodiment of the present application provides a target node, which is any node on the target path of the target business flow, including: a time slot reservation module, used to reserve a first target outgoing time slot for the target business within the scheduling cycle, wherein the number of time slot offsets of the first target outgoing time slot relative to the sending time slot is less than the number of first time slots included in the scheduling cycle; a time slot conversion module, used to convert the first target outgoing time slot into a second target outgoing time slot in the scheduling cycle based on a preset conversion strategy after receiving the forwarding message of the target business flow; a message storage module, used to store the forwarding message in the to-be-sent queue corresponding to the second target outgoing time slot for sending.
[0018] In a third aspect, an embodiment of the present application provides an electronic device, which includes a processor and a memory, wherein the memory stores at least one computer program, and the at least one computer program is loaded and executed by the processor to implement the above-mentioned message sending method.
[0019] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which at least one computer program is stored. The computer program is loaded and executed by a processor to implement the above-mentioned message sending method.
[0020] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS
[0021] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.
[0022] FIG1 is a schematic diagram showing a flow chart of a message sending method provided in an embodiment of the present application;
[0023] FIG2 shows a schematic diagram of a network structure provided by an embodiment of the present application;
[0024] FIG3 shows a schematic diagram of conversion between time slots in an arrangement cycle and time slots in a scheduling cycle according to an embodiment of the present application;
[0025] FIG4 shows a schematic diagram of the structure of an extended network protocol provided in an embodiment of the present application;
[0026] FIG5 is a schematic diagram showing the relationship between the burst volume (or sub-burst volume) and the offset of the scheduling period of the service provided in an embodiment of the present application;
[0027] FIG6 shows a schematic diagram of a time slot mapping relationship between two adjacent upstream and downstream nodes provided by an embodiment of the present application;
[0028] FIG7 shows a schematic diagram of the structure of a target node provided in an embodiment of the present application;
[0029] FIG8 shows a schematic structural diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0030] Exemplary embodiments will be described in detail herein, with examples illustrated in the accompanying drawings. In the following description, when referring to the drawings, identical numerals in different figures represent identical or similar elements, unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.
[0031] The IETF's personal draft states that at the network's ingress node, corresponding outbound time slot resources need to be reserved on the egress port for traffic flows with corresponding arrival times. At intermediate nodes in the network, corresponding outbound time slot resources need to be reserved on the egress port for traffic flows with corresponding inbound time slots. For example, taking two adjacent nodes on a traffic transmission path as an example, the outbound time slot of the upstream node's egress port is mapped to the in-progress time slot of the downstream node's egress port. Based on this mapping, outbound time slot resources can be reserved for the traffic flow on the downstream node's egress port. Forwarded messages are then sent sequentially in the outbound time slot of the egress port of each node in the deterministic path.
[0032] To facilitate understanding of the embodiments of the present application, the relevant terms are explained as follows:
[0033] Timeslot: The minimum unit of time division multiplexing. A reasonable value, such as 10us, must be designed to allow at least one complete message to be sent.
[0034] Incoming Timeslot: For an intermediate node in a specific path, the timeslot contained in the message it receives from the upstream node (i.e., the outgoing timeslot of the upstream node) is its incoming timeslot.
[0035] Outgoing Timeslot: When an intermediate node on a specific path continues to send a message to a downstream node, it selects a designated timeslot based on resource reservation or certain rules to send the message. This timeslot is called the outgoing timeslot.
[0036] Ongoing Sending Timeslot: For the head node in a path, it continues to send the message received from the client to the downstream node. When the message reaches the egress port, the time slot in which the egress port is currently sending is called the sending timeslot. For the intermediate node in a path, it continues to send the message received from the upstream node to the downstream node. When the end of the inbound timeslot to which the message belongs reaches the egress port, the time slot in which the egress port is currently sending is called the sending timeslot. Note that the sending timeslot is not an outbound timeslot.
[0037] Scheduling Period: The cyclic period of the time-division multiplexing scheduling mechanism implemented by network nodes, consisting of a fixed number of time slots. For example, a scheduling period is fixed to contain 100 time slots.
[0038] Service Burst Interval: The traffic specification for a deterministic service flow typically follows a rule that a certain burst volume is generated within a recurring burst interval (i.e., the service burst interval). For example, if a service flow generates 1000 bits of traffic every 1ms, the corresponding service burst interval is 1ms.
[0039] Orchestration Period: The control plane uses the orchestration period based on the needs of all deterministic services. It allocates time slot resources to services within the orchestration period. Specifically, it specifies which time slots a service occupies and how many bits it occupies within the orchestration period. The orchestration period is typically an integer multiple of, or equal to, the scheduling period.
[0040] Since the orchestration cycle is service-oriented, that is, the control plane determines the orchestration cycle based on the service burst intervals of all deterministic services and the discrete burst volume within each service burst interval, the service burst intervals of some deterministic services are very large. Therefore, the determined orchestration cycle length is much longer than the scheduling cycle length. If the node implements the same length scheduling cycle on the forwarding plane, it will generate a large number of queues or cache requirements, even far exceeding the node's carrying capacity. At this time, network congestion is likely to occur, seriously affecting the forwarding efficiency of the message.
[0041] Based on this, an embodiment of the present application provides a message sending method to reduce the queue or cache requirements of the target node, reduce the possibility of network congestion, and improve the forwarding efficiency of the message. Figure 1 shows a flow chart of the message sending method provided by an embodiment of the present application. As shown in the figure, the message sending method 100 may specifically include the following steps:
[0042] S101: The target node reserves a first target outgoing time slot for the target service within the orchestration cycle.
[0043] Among them, the number of time slot offsets of the first target outgoing time slot relative to the sending ongoing time slot of the target node is less than the number of first time slots included in the scheduling cycle, and the target node is any node on the target path of the target service flow.
[0044] S102: After receiving the forwarding message of the target service flow, the target node converts the first target outgoing time slot into a second target outgoing time slot in the scheduling period based on a preset conversion strategy;
[0045] S103: The target node stores the forwarded message in a waiting queue corresponding to the second target outgoing time slot for sending.
[0046] Through the above steps, the target node first reserves a first target outbound time slot for the target service within the scheduling cycle, and the number of time slot offsets of the first target outbound time slot relative to the target node's sending time slot is less than the number of first time slots included in the scheduling cycle. That is, in the process of reserving the outbound time slot, the target node not only considers the number of time slot offsets of the target node's outbound time slot relative to the sending time slot, and satisfies the residence delay evaluation value determined according to the number of time slot offsets to meet the residence delay budget value corresponding to the target node, but also further considers the number of time slots in the scheduling cycle, and the number of time slot offsets is less than the number of time slots in the scheduling cycle.
[0047] Furthermore, after receiving the forwarding message of the target business flow, the target node converts the first target outgoing time slot into the second target outgoing time slot in the scheduling cycle based on the preset conversion strategy; and stores the forwarding message in the waiting queue corresponding to the second target outgoing time slot for sending. In this way, even if the service burst interval is very large, making the length of the orchestration cycle much longer than the length of the scheduling cycle, the queues maintained on the target node are still very small, thereby reducing the cache requirements of the target node and improving the forwarding efficiency of the message.
[0048] In a possible implementation, in S102, after receiving the forwarding message of the target business flow, the target node converts the first target outgoing time slot into the second target outgoing time slot in the scheduling period based on a preset conversion strategy, including: when the target node receives the forwarding message of the target business flow, obtaining the first target outgoing time slot corresponding to the forwarding message; based on the preset conversion strategy, converting the first target outgoing time slot into the second target outgoing time slot in the scheduling period.
[0049] This will be explained with reference to specific examples. Figure 2 shows a schematic diagram of a network structure provided by an embodiment of the present application. As shown in the figure, the network includes six nodes: Node S, Node A, Node C, Node E, Node D, and Node B. The links between any two nodes are bidirectional links. Within the control plane, a centralized controller or the head node of the path can calculate the end-to-end target path for the target service flow. Specifically, the target path of the target business flow can be determined in the following manner: obtain the delay requirement of the target business flow and the time slot resource information of each egress port of each node; determine at least one candidate path corresponding to the target business flow based on the time slot resource information of each egress port of each node; reserve time slot resources for the target business flow at the egress port of the candidate node in each candidate path, and determine the total residence delay evaluation value of the target business flow on the candidate path; then, select the candidate path that meets the first condition as the target path of the target business flow, the first condition including that time slot resources are successfully reserved for the target business flow at the egress ports of all candidate nodes constituting the selected candidate path, and the sum of the total residence delay evaluation value of the selected candidate path and the accumulated link propagation delay meets the delay requirement.
[0050] After the target path is successfully calculated and validated, if the target path is path_SD, the forwarded message of the target service flow is forwarded along this target path_SD via SACED. Upon receiving a forwarded message for the target service flow, any node on the target path obtains the first target outbound timeslot corresponding to the forwarded message, which is pre-reserved for the target service flow. Based on a preset conversion policy, the node converts the first target outbound timeslot into the second target outbound timeslot in the scheduling period.
[0051] The following two methods may be used to obtain the first target outgoing time slot corresponding to the forwarded message.
[0052] The first implementation method: based on the time slot number carried by the forwarding message, determine the first target outgoing time slot reserved by the target node for the target service flow within the orchestration period, and the time slot number corresponds to the first target outgoing time slot.
[0053] The second implementation method: based on the pre-stored message forwarding table entry, determine the outgoing time slot corresponding to the forwarded message; determine the outgoing time slot as the first target outgoing time slot reserved by the target node for the target service flow within the orchestration cycle.
[0054] The scheduling period implemented by the target node on the forwarding plane is not completely equivalent to the orchestration period of the control plane. Since the orchestration period is business-oriented, the length of the orchestration period is often much longer than the length of the scheduling period. Therefore, a corresponding conversion strategy is used to convert the first target outgoing time slot into the second target outgoing time slot in the scheduling period, store the message in the queue or cache corresponding to the second target outgoing time slot, and finally send the message in the second target outgoing time slot. In one possible embodiment, the preset conversion strategy includes: when the number of second time slots included in the orchestration period is an integer multiple of the number of first time slots included in the scheduling period, the first target outgoing time slot performs a modulo operation on the first number of time slots to obtain the second target outgoing time slot. That is, when the number of second time slots included in the orchestration period is an integer multiple of the number of first time slots included in the scheduling period, the time slot conversion is performed according to the following formula: Second target outgoing time slot = first target outgoing time slot % M;
[0055] Wherein, % is the remainder operation, and M is the number of time slots included in the scheduling cycle.
[0056] The explanation is given with reference to specific examples. FIG3 shows a schematic diagram of the conversion between the time slots in the scheduling cycle and the time slots in the scheduling cycle provided by an embodiment of the present application. As shown in the figure, the number of time slots N included in the scheduling cycle of the output port of node R1 (which is connected to node R2) is an integer multiple of the number of time slots M included in the scheduling cycle, that is, the scheduling cycle contains multiple rounds of scheduling cycles. The time slots from time slot 0 to time slot M-1 in the scheduling cycle are the first round of scheduling cycles, and the time slots from time slot M to time slot 2M-1 are the second round of scheduling cycles. By analogy, the time slots from time slot NM to time slot N-1 are the N / Mth round of scheduling cycles. Each time slot in the scheduling cycle corresponds to an associated queue (or buffer) for storing messages to be sent in the corresponding time slot. In this way, even if the service burst interval is very large and the length of the scheduling cycle is much greater than the length of the scheduling cycle, the queues maintained on the target node are still very small, thereby reducing the cache requirements of the target node and improving the forwarding efficiency of the message.
[0057] The following describes a process in which the control plane determines a target path that meets a specific delay requirement for a target service flow and reserves a target time slot on an egress port along the target path.
[0058] As mentioned earlier, within the control plane, the centralized controller or the head node of the path calculates the end-to-end target path for the target service flow. Time slot resources are successfully reserved for the target service flow on the egress ports of all target nodes of the target path, and the sum of the total residence delay evaluation value of the selected candidate path and the accumulated link propagation delay meets the delay requirement.
[0059] Taking the network structure in Figure 2 as an example, assume the end-to-end delay required by the target service flow is 200µs. The centralized controller or head node S first identifies a candidate path, path_SD, passing through node SACED. The accumulated link propagation delay of this candidate path is: 20µs + 10µs + 10µs + 10µs = 50µs. The difference between the target service flow's delay requirement and the accumulated link propagation delay is used to determine the total residence delay budget for each candidate node in the candidate path: 200µs - 50µs = 150µs. Assuming a mean budgeting approach, each candidate node will share a residence delay budget of 150 / 5 = 30µs.
[0060] Along the candidate path path_SD from the head node S to the tail node D, time slot resources are reserved for the target business flow at the egress port of each candidate node in turn. For each candidate node, the target time slot is reserved for the target business flow based on the time slot resource information of the egress port of the candidate node and the residence delay budget value of the candidate node determined by the above steps, and the residence delay evaluation value corresponding to the candidate node is determined. Then, when time slot resources are successfully reserved for the target business flow at the egress ports of all candidate nodes in the candidate path path_SD, and the sum of the total residence delay evaluation value of the selected candidate path and the accumulated link propagation delay meets the delay requirement of the target business flow, the candidate path path_SD is determined as the target path of the target business flow, and all candidate nodes contained in the target path are determined as target nodes. At this time, the path path_SD is calculated successfully and takes effect.
[0061] In a possible implementation, the target node reserves a first target outgoing time slot for the target service within the orchestration period, including:
[0062] Step 1011: Acquire the time slot resource information of the target node in the scheduling period and the number of first time slots included in the scheduling period of the target node;
[0063] Step 1012: Based on the predetermined residence delay budget value corresponding to the target node and the time slot resource information, reserve a target time slot for the target service, and determine a residence delay evaluation value corresponding to the target node;
[0064] Step 1013: When the difference between the resident delay budget value and the resident delay evaluation value is less than a preset threshold, and the number of time slot offsets of the target time slot relative to the sending-in-progress time slot of the target node is less than the first number of time slots, the target time slot is determined as the first target outbound time slot reserved by the target node for the target service within the scheduling cycle; wherein the sending-in-progress time slot is used to indicate a time slot in a sending state on the egress port of the target node.
[0065] Through the above steps, since the scheduling cycle of the target node is much longer than the scheduling cycle, when the target node reserves the target time slot for the target service flow, in addition to satisfying that the difference between the residence delay budget value and the residence delay evaluation value is less than the preset threshold, the target node also needs to satisfy that the number of time slot offsets of the target time slot relative to the target node's sending time slot is less than the first time slot number.
[0066] The target nodes may notify each other of the time slot resource information of the outbound port and the first number of time slots included in the scheduling period. That is, in step 1011, obtaining the time slot resource information of the target node in the scheduling period and the first number of time slots included in the scheduling period of the target node includes:
[0067] The target node is notified of time slot resource information within the scheduling period and length information of the scheduling period through an extended network protocol; the length information is used to indicate the number of first time slots included in the scheduling period or the length of the scheduling period.
[0068] In a specific implementation, the IGP protocol (Interior Gateway Protocol), OSPF (Open Shortest Path First) or other similar protocols can be extended, and the BGP-LS protocol (Border Gateway Protocol-Link State) or other similar protocols can also be extended to support the notification of the delay resource information of the outbound port. The ISIS protocol suite (refer to RFC5305, RFC5311, RFC5316, RFC5120) has defined TLVs22 / 23 / 141 / 222 / 223 for notifying adjacency relationships. The embodiment of the present application adds a new Sub-TLV to these TLVs, denoted as Timeslot Scheduling Capability Sub-TLV, which is used to carry the time slot resource information of the outbound port and the number of the first time slots included in the scheduling cycle. Figure 4 shows a structural diagram of an extended network protocol provided by an embodiment of the present application. As shown in the figure, the extended network protocol includes the following fields:
[0069] Type: occupies 1 byte. Its value is to be assigned by IANA (Internet Assigned Numbers Authority) to indicate that this TLV is the Timeslot Scheduling Capability Sub-TLV.
[0070] Length: 1 byte, indicating the content length of this TLV (excluding the Type and Length fields themselves).
[0071] Timeslot Count of OP: occupies 2 bytes and indicates the number of time slots (N) contained in the orchestration period. These time slots are numbered in sequence from 0 to N-1.
[0072] Timeslot Count of SP: occupies 2 bytes and indicates the number of time slots (M) contained in the scheduling period. These time slots are numbered from 0 to M-1.
[0073] Timeslot Interval: occupies 2 bytes and indicates the length of each timeslot in microseconds (us).
[0074] It should be noted that in other extension modes, the time slot length may be expressed in milliseconds (ms) or nanoseconds (ns), or the length of the scheduling cycle or the scheduling cycle may be represented in time units.
[0075] By parsing the extended network protocol, the time slot resource information of the target node in the scheduling period and the number of first time slots included in the scheduling period of the target node are obtained.
[0076] Furthermore, based on the predetermined residence delay budget value corresponding to the target node and the time slot resource information, a target time slot is reserved for the target service, and the residence delay evaluation value corresponding to the target node is determined. To illustrate with a specific example, FIG5 shows a schematic diagram of the offset relationship between the burst volume (or sub-burst volume) and the scheduling period of the service provided in an embodiment of the present application. As shown in the figure, when the target node is the head node, the head node H reserves a target time slot within the scheduling period for each target burst volume contained in the target service flow when receiving the target service flow. Here, the target service flow is a periodic traffic flow, and its service burst interval is equal to the length of the scheduling period of the head node after expansion, so that the head node receives the same target service flow within each scheduling period. In the figure, it is assumed that the target service flow has n target burst volumes distributed within each scheduling period, which are recorded as burst_1, burst_2, ..., burst_n. The moment a specific burst arrives at the ingress port of the head node H, undergoes an intra-node forwarding delay d_f, and finally reaches the egress port (link-a) is offset from the orchestration period of that egress port. For example, when burst burst_i arrives at the egress port, the egress port's ongoing transmission slot is y, and the remaining time until the end of slot y is remaining_y.
[0077] Then, when the difference between the dwell time budget value and the dwell time evaluation value is less than a preset threshold, and the number of slot offsets of the target slot relative to the ongoing transmission slot of the target node is less than the number of the first slots, the outgoing slot reserved by the head node H for burst_i is offset by O (≥1) slots after its corresponding ongoing transmission slot y, that is, the number of the first target outgoing slot is (y + O) % N, where N is the number of the first slots included in the scheduling period, and % is the modulo operator.
[0078] FIG. 6 shows a schematic diagram of the slot mapping relationship between two adjacent upstream and downstream nodes provided by an embodiment of the present application. As shown in the figure, when the target node is the intermediate node I, when the intermediate node I on the path receives the target traffic flow sent by the upstream node U, it reserves target slots for each burst volume included in the target traffic flow within the scheduling period. Similarly, the service burst interval is extended to be equal to the length of the scheduling period of the intermediate node I, so that the intermediate node I receives the same target traffic flow in each scheduling period. In the figure, it is assumed that the upstream node U sends the burst volume burst_i in its outgoing slot x, and the mapping relationship between the outgoing slot x of the outgoing port link-a of the upstream node U and the ongoing transmission slot y of the outgoing port link-b of the intermediate node I can be obtained. This mapping relationship means that when the end of the outgoing slot x of the upstream node U reaches the incoming port of the intermediate node I and passes through an in-node forwarding delay d_f and then reaches the outgoing port (link-b), the ongoing transmission slot of the outgoing port is y, and the remaining time until the end of the slot y is remaining_xy.
[0079] Then, when the difference between the dwell time budget value and the dwell time evaluation value is less than a preset threshold, and the number of slot offsets of the target slot relative to the ongoing transmission slot of the target node is less than the number of slots, the outgoing slot reserved by the intermediate node I for burst_i is offset by O (≥1) slots after its corresponding ongoing transmission slot y, that is, the number of the first target outgoing slot is (y + O) % N, where N is the number of slots included in the scheduling period, and % is the modulo operator.
[0080] According to the above slot resource reservation process, when the target burst volume corresponding to the first target outgoing slot (z) reaches the outgoing port of the node R1, the first ongoing transmission slot (y) in the scheduling period of the outgoing port at this time must be offset by O before the first target outgoing slot, and O < M, which means that the target burst volume does not arrive at this node randomly, but strictly follows the time so that it will definitely fall into the first ongoing transmission slot (y) when it reaches the outgoing port.
[0081] Next, we demonstrate that the target burst arriving at the egress port during the first transmit-in-progress time slot (y) can be safely inserted into the corresponding queue in the scheduling cycle without queue overflow. The demonstration process is as follows:
[0082] Assume that each time slot in the scheduling cycle has a virtual queue, and the length of the virtual queue is the initial bit size corresponding to the time slot. For example, if the virtual queue corresponding to the first target outgoing time slot (z) is queue-z, then the packets that can be inserted into queue-z can only come from the following burst sizes:
[0083] The burst volume arriving at the egress port during the sending time slot y=(z-M+1+N)%N, that is, the first target outgoing time slot (z) is reserved for this type of burst volume according to O=M-1.
[0084] Or the burst volume arriving at the egress port during the sending in progress time slot y=(z-M+2+N)%N, that is, the first target outgoing time slot (z) is reserved for this type of burst volume according to O=M-2.
[0085] And so on.
[0086] Or the burst volume arriving at the egress port during the sending in the time slot y=(z-1+N)%N, that is, the first target outgoing time slot (z) is reserved according to O=1 for such burst volume.
[0087] The reserved sum of all these burst sizes does not exceed the initial bit size of the first target outgoing time slot (z).
[0088] Then, when the sending time slot becomes z, queue-z will be sent and completely cleared. After that, in the scheduling cycle, starting from time slot z+1 until the last time slot N-1, queue-z will no longer have any messages inserted. Obviously, such a virtual queue design greatly wastes resources. In fact, queue-z can be reused by the subsequent first target outgoing time slot (z+M)%N. That is:
[0089] The burst volume arriving at the egress port during the sending in the time slot y=(z+1)%N is a burst volume. This type of burst volume can reserve the first target outgoing time slot (z+M)%N according to O=M-1.
[0090] Or the burst volume that arrives at the egress port during the sending in-progress time slot y=(z+2)%N, such burst volume reserves the first target outgoing time slot (z+M)%N according to O=M-2.
[0091] And so on.
[0092] Or the burst volume arriving at the egress port during the sending time slot y=(z+M-1)%N, that is, the first target outgoing time slot (z+M)%N is reserved according to O=1 for this type of burst volume.
[0093] The total reserved sum of all these bursts does not exceed the initial bit amount of the first target outgoing time slot (z + M).
[0094] It is not difficult to know that queue-z can be used by any target outgoing time slot (z + k*M) % N, where k is a non-negative integer. By observing (z + k*M) % N, it can be seen that the smallest z satisfies 0 ≤ z < M. That is, in fact, only M queues are needed to store packets during the entire scheduling period. These M queues are the queues corresponding to the M time slots in the scheduling period. That is to say, the smallest z (satisfying 0 ≤ z < M) is the number of the time slot in the scheduling period, and the target outgoing time slot (z + k*M) % N is the number of the time slot in the scheduling period. The result of taking the modulo of the latter by M is the former, and then the queue corresponding to the former can be accessed. In short, the reason why a queue can store packets of multiple target outgoing time slots without overflow is that, as analyzed above, the packets stored in this queue before M time slots have been sent.
[0095] From the above demonstration process, it can be seen that by adopting the method provided in the embodiment of the present application, when the length of the scheduling period is much greater than the length of the scheduling period, the queues maintained on the target node are still very few, so that the cache requirements of the target node can be significantly reduced, the possibility of network congestion can be reduced, and the forwarding efficiency of packets can be improved.
[0096] FIG. 7 shows a schematic structural diagram of a target node provided in an embodiment of the present application. As shown in the figure, the target node 700 includes: a time slot reservation module 710, configured to reserve a first target outgoing time slot for a target service during a scheduling period, where the number of time slot offsets of the first target outgoing time slot relative to the in-progress sending time slot is less than the number of the first time slots included in the scheduling period; a time slot conversion module 720, configured to convert the first target outgoing time slot into a second target outgoing time slot in the scheduling period based on a preset conversion strategy after receiving a forwarding packet of the target service flow; and a packet storage module 730, configured to store the forwarding packet in a queue to be sent corresponding to the second target outgoing time slot for sending.
[0097] In a possible implementation manner, the time slot conversion module 720 includes: a time slot acquisition unit, configured to acquire the first target outgoing time slot corresponding to the forwarding packet when receiving the forwarding packet of the target service flow; and a time slot conversion unit, configured to convert the first target outgoing time slot into a second target outgoing time slot in the scheduling period based on a preset conversion strategy.
[0098] In one possible embodiment, when the time slot acquisition unit is used to obtain the first target outgoing time slot corresponding to the forwarded message, it is specifically used to: determine the first target outgoing time slot reserved by the target node for the target service flow within the orchestration cycle based on the time slot number carried by the forwarded message, and the time slot number corresponds to the first target outgoing time slot.
[0099] In one possible implementation, when the time slot acquisition unit is used to obtain the first target outgoing time slot corresponding to the forwarded message, it is specifically used to: determine the outgoing time slot corresponding to the forwarded message based on a pre-stored message forwarding table entry; and determine the outgoing time slot as the first target outgoing time slot reserved by the target node for the target service flow within the orchestration cycle.
[0100] The preset conversion strategy includes: when the number of second time slots included in the scheduling period is an integer multiple of the number of first time slots included in the scheduling period, the first target outgoing time slot performs a modulo operation on the first number of time slots to obtain the second target outgoing time slot.
[0101] In one possible embodiment, the time slot reservation module 710 includes: an information acquisition unit, configured to acquire time slot resource information of the target node within the scheduling period, and a first number of time slots included in the scheduling period of the target node; a determination unit, configured to reserve a target time slot for the target service and determine a resident delay evaluation value corresponding to the target node based on a predetermined resident delay budget value corresponding to the target node and the time slot resource information; a reservation unit, configured to determine the target time slot as the first target outbound time slot reserved by the target node for the target service within the scheduling period, if a difference between the resident delay budget value and the resident delay evaluation value is less than a preset threshold and the number of time slot offsets of the target time slot relative to the target node's sending-in-progress time slot is less than the first number of time slots; wherein the sending-in-progress time slot is used to indicate a time slot in a sending state on the egress port of the target node.
[0102] In one possible implementation, when the information acquisition unit is used to acquire the time slot resource information of the target node within the scheduling period and the first number of time slots included in the scheduling period of the target node, it is specifically used to: notify the time slot resource information of the target node within the scheduling period and the length information of the scheduling period through an extended network protocol; the length information is used to indicate the first number of time slots included in the scheduling period or the length of the scheduling period.
[0103] FIG8 shows a schematic diagram of the hardware structure of an electronic device that implements an embodiment of the present application. Referring to the figure, at the hardware level, the electronic device includes a processor, and optionally, an internal bus, a network interface, and a memory. The memory may include a memory, such as a high-speed random access memory (RAM), and may also include a non-volatile memory (non-volatile memory), such as at least one disk storage device. Of course, the computer device may also include hardware required for other services.
[0104] The processor, network interface, and memory can be interconnected via an internal bus, such as an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus. These buses can be classified as address buses, data buses, and control buses. For ease of illustration, the figure uses only one bidirectional arrow, but this does not imply that there is only one bus or only one type of bus.
[0105] The memory stores programs. Specifically, the programs may include program code, which includes computer operating instructions. The memory may include internal memory and non-volatile memory, and provides instructions and data to the processor.
[0106] The processor reads the corresponding computer program from the non-volatile memory into the internal memory and then runs it, logically forming a device for locating the target user. The processor executes the program stored in the memory and specifically performs the method disclosed in the embodiment shown in Figure 1 and implements the functions and benefits of the various methods described in the method embodiments above, which will not be repeated here.
[0107] The method disclosed in the embodiment shown in FIG1 of the present application can be implemented in a processor or by a processor. The processor may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method can be completed by hardware integrated logic circuits in the processor or software instructions. The above processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The various methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in the embodiments of the present application can be directly embodied as being executed by a hardware decoding processor, or can be executed by a combination of hardware and software modules in the decoding processor. The software module can be located in a storage medium well-known in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, etc. The storage medium is located in the memory, and the processor reads the information in the memory and, in conjunction with its hardware, completes the steps of the above method.
[0108] The computer device can also execute the methods described in the above method embodiments and realize the functions and beneficial effects of the methods described in the above method embodiments, which will not be repeated here.
[0109] Of course, in addition to software implementation, the electronic device of this application does not exclude other implementation methods, such as logic devices or a combination of software and hardware, etc. That is to say, the execution subject of the following processing flow is not limited to each logic unit, but can also be hardware or logic devices.
[0110] An embodiment of the present application also proposes a computer-readable storage medium, which stores one or more programs. When the one or more programs are executed by an electronic device including multiple applications, the electronic device executes the method disclosed in the embodiment shown in Figure 1 and realizes the functions and beneficial effects of the various methods described in the previous method embodiments, which will not be repeated here.
[0111] The computer-readable storage medium includes a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.
[0112] Furthermore, an embodiment of the present application also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the following process is implemented: the method disclosed in the embodiment shown in Figure 1 and the functions and beneficial effects of the various methods described in the previous method embodiments are implemented, which will not be repeated here.
[0113] In short, the above description is only a preferred embodiment of the present application and does not limit the scope of protection of the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.
[0114] The systems, devices, modules, or units described in the above embodiments may be implemented by computer chips or entities, or by products having certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.
[0115] Computer-readable media include permanent and non-permanent, removable and non-removable media that can be used to store information by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change random access memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can store information that can be accessed by a computing device. According to the definition in this article, computer-readable media does not include transitory media such as modulated data signals and carrier waves.
[0116] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.
[0117] The various embodiments in this specification are described in a progressive manner. Similar parts between the various embodiments can be referred to in conjunction with each other. Each embodiment focuses on the differences between the other embodiments. In particular, the system embodiments are generally similar to the method embodiments, so the description is relatively simple. For relevant parts, refer to the description of the method embodiments.
Claims
1. A message sending method, comprising: The target node reserves a first target outgoing time slot for the target service within the scheduling period, wherein the number of time slot offsets of the first target outgoing time slot relative to the sending-in-progress time slot of the target node is less than the number of first time slots included in the scheduling period, and the target node is any node on the target path of the target service flow; After receiving the forwarding message of the target service flow, the target node converts the first target outgoing time slot into a second target outgoing time slot in the scheduling period based on a preset conversion strategy; The target node stores the forwarded message in a to-be-sent queue corresponding to the second target outgoing time slot for sending.
2. The method according to claim 1, wherein: After receiving the forwarding message of the target service flow, the target node converts the first target outgoing time slot into the second target outgoing time slot in the scheduling period based on a preset conversion strategy, including: When receiving the forwarding message of the target service flow, the target node obtains the first target outgoing time slot corresponding to the forwarding message; based on a preset conversion strategy, the first target outgoing time slot is converted into a second target outgoing time slot in the scheduling period.
3. The method according to claim 2, wherein: The obtaining of the first target outgoing time slot corresponding to the forwarded message includes: Based on the time slot number carried by the forwarding message, a first target outbound time slot reserved by the target node for the target service flow within the orchestration period is determined, and the time slot number corresponds to the first target outbound time slot.
4. The method according to claim 2, wherein: The obtaining of the first target outgoing time slot corresponding to the forwarded message includes: Determining an outgoing time slot corresponding to the forwarded message based on a pre-stored message forwarding table entry; The outgoing time slot is determined as the target node's time slot for the target service flow in the scheduling cycle. The first target outgoing time slot reserved.
5. The method according to claim 1, wherein: The preset conversion strategy includes: When the number of second time slots included in the scheduling period is an integer multiple of the number of first time slots included in the scheduling period, the first target outbound time slot performs a modulo operation on the first number of time slots to obtain the second target outbound time slot.
6. The method according to claim 1, wherein: The target node reserves a first target outgoing time slot for the target service within the scheduling period, including: Acquire time slot resource information of the target node in a scheduling period, and the number of first time slots included in the scheduling period of the target node; Based on the predetermined resident delay budget value corresponding to the target node and the time slot resource information, reserve a target time slot for the target service, and determine a resident delay evaluation value corresponding to the target node; When the difference between the resident delay budget value and the resident delay evaluation value is less than a preset threshold, and the number of time slot offsets of the target time slot relative to the sending time slot of the target node is less than the first number of time slots, the target time slot is determined as the first target outbound time slot reserved by the target node for the target service within the scheduling cycle; wherein the sending time slot is used to indicate a time slot in a sending state on the egress port of the target node.
7. The method according to claim 6, wherein: The acquiring the time slot resource information of the target node in the scheduling period and the number of first time slots included in the scheduling period of the target node includes: The target node is notified of time slot resource information within the scheduling period and length information of the scheduling period through an extended network protocol; the length information is used to indicate the number of first time slots included in the scheduling period or the length of the scheduling period.
8. A target node, the target node being any node on a target path of a target service flow, comprising: The time slot reservation module is used to reserve the first target outgoing time slot for the target service within the scheduling cycle. middle, The number of time slot offsets of the first target outgoing time slot relative to the sending-in-progress time slot is less than the number of first time slots included in the scheduling cycle; A time slot conversion module, configured to convert the first target outgoing time slot into a second target outgoing time slot in the scheduling period based on a preset conversion strategy after receiving a forwarding message of the target service flow; The message storage module is used to store the forwarded message in the waiting-to-send queue corresponding to the second target outgoing time slot for sending.
9. An electronic device, comprising a processor and a memory, wherein the memory stores a program or instruction that can be run on the processor, and when the program or instruction is executed by the processor, the steps of the method according to any one of claims 1 to 7 are implemented.
10. A readable storage medium storing a program or an instruction, wherein the program or the instruction is executed by a processor to implement the steps of the method according to any one of claims 1 to 7.
Citation Information
Patent Citations
Message forwarding method, electronic equipment and storage medium
CN115037399A
Distributed rescheduling of bounded flows in a time sensitive network
US20160021017A1
