Search

CN-122027572-A - Traffic isolation method and device, electronic equipment and storage medium

CN122027572ACN 122027572 ACN122027572 ACN 122027572ACN-122027572-A

Abstract

The disclosure provides a flow isolation method, a flow isolation device, electronic equipment and a storage medium, and relates to the field of artificial intelligence such as cloud computing and flow management. The method can include the steps of responding to determining that a scheduling triggering condition is met, determining target equipment group nodes to be scheduled from all equipment group nodes according to node configuration information corresponding to the equipment group nodes respectively, wherein the node configuration information comprises rate constraint information, determining target equipment nodes to be scheduled from all equipment nodes according to node configuration information of all equipment nodes subordinate to the target equipment group nodes, determining the target equipment nodes as target scheduling results, and enabling all the equipment group nodes to respectively belong to at least one equipment node for transmitting the same type of traffic, wherein the number of the types is M, the M is a positive integer larger than 1, and each type corresponds to at least one equipment group node respectively. By applying the scheme disclosed by the disclosure, the resource utilization efficiency, the rationality of resource utilization and the like can be improved.

Inventors

  • TAN YITAO

Assignees

  • 北京百度网讯科技有限公司

Dates

Publication Date
20260512
Application Date
20260123

Claims (15)

  1. 1. A method of traffic isolation, comprising: Responding to the determination that the scheduling triggering condition is met, and determining target equipment group nodes to be scheduled from all the equipment group nodes according to node configuration information corresponding to the node configuration information, wherein the node configuration information comprises rate constraint information; And determining target equipment nodes to be scheduled from the equipment nodes according to the node configuration information of the equipment nodes subordinate to the target equipment group nodes, determining the target equipment nodes as target scheduling results, wherein the equipment nodes subordinate to the equipment nodes respectively belong to at least one equipment node for transmitting the same type of traffic, the number of the types is M, the M is a positive integer greater than 1, and each type corresponds to at least one equipment group node respectively.
  2. 2. The method of claim 1, wherein, The equipment group node is positioned in an equipment group scheduling layer, and the equipment node is positioned in the equipment scheduling layer; The layering in the layering service quality scheduling system also comprises a global scheduling layer and a virtual machine scheduling layer; each virtual machine node in the virtual machine scheduling layer is a subordinate node of the global node in the global scheduling layer, each virtual machine node is respectively subordinate to M equipment group nodes, and the M equipment group nodes respectively correspond to different types of traffic.
  3. 3. The method of claim 2, wherein, The value of M is 2, and the different types of traffic comprise remote direct memory access traffic and transmission control protocol traffic.
  4. 4. The method of claim 2, wherein, The rate constraint information comprises a basic promised speed limit and a maximum burst speed limit; The node configuration information also comprises node priority.
  5. 5. The method of claim 4, wherein, Responsive to not supporting a virtual machine burst, the sum of the basic committed speed limits of each virtual machine node and the sum of the maximum burst speed limits of each virtual machine node are both equal to a global speed limit of the global node; And the sum of the basic promised speed limits of all the equipment group nodes belonging to the same virtual machine node is equal to the basic promised speed limit of the virtual machine node to which the equipment group nodes belong, and the maximum burst speed limit of all the equipment group nodes belonging to the same virtual machine node is equal to the maximum burst speed limit of the virtual machine node to which the equipment group nodes belong; And the basic commitment speed limit of each device node belonging to the same device group node is equal to the basic commitment speed limit of the device group node to which the device node belongs, and the maximum burst speed limit of each device node belonging to the same device group node is equal to the maximum burst speed limit of the device group node to which the device node belongs.
  6. 6. The method of claim 4, wherein, In response to supporting virtual machine bursting, the sum of the basic committed speed limits of each virtual machine node is equal to a global speed limit of the global node, and the maximum burst speed limit of each virtual machine node is equal to the global speed limit; And the sum of the basic promised speed limits of all the equipment group nodes belonging to the same virtual machine node is equal to the basic promised speed limit of the virtual machine node to which the equipment group nodes belong, and the maximum burst speed limit of all the equipment group nodes belonging to the same virtual machine node is equal to the maximum burst speed limit of the virtual machine node to which the equipment group nodes belong; And the basic commitment speed limit of each device node belonging to the same device group node is equal to the basic commitment speed limit of the device group node to which the device node belongs, and the maximum burst speed limit of each device node belonging to the same device group node is equal to the maximum burst speed limit of the device group node to which the device node belongs.
  7. 7. The method of claim 4, wherein, The determining the target equipment group node to be scheduled from the equipment group nodes comprises the following steps: Determining target virtual machine nodes to be scheduled from the virtual machine nodes according to the node configuration information of the virtual machine nodes; And determining the target equipment group node from the equipment group nodes subordinate to the target virtual machine node according to the node configuration information of the equipment group nodes subordinate to the target virtual machine node.
  8. 8. The method of claim 7, wherein, Determining a target virtual machine node to be scheduled from all virtual machine nodes, wherein determining a schedulable virtual machine node from all virtual machine nodes, determining the target virtual machine node from the schedulable virtual machine nodes, and the schedulable virtual machine node comprises the maximum burst speed limit of which the speed does not exceed the maximum burst speed limit; determining the target equipment group node from all the equipment group nodes subordinate to the target virtual machine node comprises determining a schedulable equipment group node from all the equipment group nodes subordinate to the target virtual machine node, and determining the target equipment group node from the schedulable equipment group node; Determining the target equipment node to be scheduled from the equipment nodes according to the node configuration information of the equipment nodes subordinate to the target equipment group node comprises determining the schedulable equipment node from the equipment nodes subordinate to the target equipment group node and determining the target equipment node from the schedulable equipment node.
  9. 9. The method of claim 8, wherein, Determining the target virtual machine node from the schedulable virtual machine nodes comprises the steps of determining the scheduling priority of each schedulable virtual machine node respectively, and determining the schedulable virtual machine node with the highest scheduling priority as the target virtual machine node; Determining the target equipment group node from the schedulable equipment group nodes comprises the steps of determining the scheduling priority of each schedulable equipment group node respectively, and determining the schedulable equipment group node with the highest scheduling priority as the target equipment group node; Determining the target equipment node from the schedulable equipment nodes comprises the steps of determining the scheduling priority of each schedulable equipment node respectively, and determining the schedulable equipment node with the highest scheduling priority as the target equipment node.
  10. 10. The method of claim 9, wherein, The node priority comprises a high priority and a low priority; the scheduling priorities include, in order from high to low, a first priority, a second priority, a third priority, and a fourth priority.
  11. 11. The method of claim 10, wherein, For any node to be processed, the node to be processed is the schedulable virtual machine node, the schedulable equipment group node or the schedulable equipment node, and the method for determining the scheduling priority of the node to be processed includes: In response to determining that the rate of the node to be processed does not exceed the base committed speed limit of itself, and determining that the node to be processed is the high priority, determining that the scheduling priority is the first priority; responsive to determining that the rate of the node to be processed does not exceed the base committed speed limit of itself, and determining that the node to be processed is the low priority, determining that the scheduling priority is the second priority; Responsive to determining that the rate of the node to be processed exceeds the base committed rate limit of itself, but does not exceed the maximum bursty rate limit of itself, and determining that the node to be processed is the high priority, determining that the scheduling priority is the third priority; And in response to determining that the rate of the node to be processed exceeds the base committed rate limit of itself but does not exceed the maximum burst rate limit of itself, and determining that the node to be processed is the low priority, determining that the scheduling priority is the fourth priority.
  12. 12. A traffic isolation device comprises a first scheduling module and a second scheduling module; the first scheduling module is used for determining target equipment group nodes to be scheduled from all the equipment group nodes according to node configuration information corresponding to the first scheduling module in response to determining that the scheduling triggering conditions are met, and the node configuration information comprises rate constraint information; The second scheduling module is configured to determine a target device node to be scheduled from the device nodes according to the node configuration information of the device nodes subordinate to the target device group node, determine the target device node as a target scheduling result, and each device group node belongs to at least one device node for transmitting the same type of traffic, where the number of types is M, M is a positive integer greater than 1, and each type corresponds to at least one device group node.
  13. 13. An electronic device, comprising: at least one processor, and A memory communicatively coupled to the at least one processor, wherein, The memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-11.
  14. 14. A non-transitory computer readable storage medium storing computer instructions for causing a computer to perform the method of any one of claims 1-11.
  15. 15. A computer program product comprising computer programs/instructions which, when executed by a processor, implement the method of any of claims 1-11.

Description

Traffic isolation method and device, electronic equipment and storage medium Technical Field The disclosure relates to the technical field of artificial intelligence, in particular to the fields of cloud computing, flow management and the like, and particularly relates to a flow isolation method, a flow isolation device, electronic equipment and a storage medium. Background Remote direct memory access (RDMA, remote Direct Memory Access) traffic and transmission control protocol (TCP, transmission Control Protocol) traffic can be supported simultaneously in the virtual private cloud (VPC, virtual Private Cloud), which brings advantages to both the tenant and cloud provider, such as meeting the flexible demands of the tenant for RDMA traffic and TCP traffic, and in addition, compared to building two independent overlay networks, the implementation and operation costs are reduced. Disclosure of Invention The disclosure provides a traffic isolation method, a traffic isolation device, electronic equipment and a storage medium. A method of traffic isolation, comprising: Responding to the determination that the scheduling triggering condition is met, and determining target equipment group nodes to be scheduled from all the equipment group nodes according to node configuration information corresponding to the node configuration information, wherein the node configuration information comprises rate constraint information; And determining target equipment nodes to be scheduled from the equipment nodes according to the node configuration information of the equipment nodes subordinate to the target equipment group nodes, determining the target equipment nodes as target scheduling results, wherein the equipment nodes subordinate to the equipment nodes respectively belong to at least one equipment node for transmitting the same type of traffic, the number of the types is M, the M is a positive integer greater than 1, and each type corresponds to at least one equipment group node respectively. A traffic isolation device comprises a first scheduling module and a second scheduling module; the first scheduling module is used for determining target equipment group nodes to be scheduled from all the equipment group nodes according to node configuration information corresponding to the first scheduling module in response to determining that the scheduling triggering conditions are met, and the node configuration information comprises rate constraint information; The second scheduling module is configured to determine a target device node to be scheduled from the device nodes according to the node configuration information of the device nodes subordinate to the target device group node, determine the target device node as a target scheduling result, and each device group node belongs to at least one device node for transmitting the same type of traffic, where the number of types is M, M is a positive integer greater than 1, and each type corresponds to at least one device group node. An electronic device, comprising: at least one processor, and A memory communicatively coupled to the at least one processor, wherein, The memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method as described above. A non-transitory computer readable storage medium storing computer instructions for causing a computer to perform a method as described above. A computer program product comprising computer programs/instructions which when executed by a processor implement a method as described above. It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the disclosure, nor is it intended to be used to limit the scope of the disclosure. Other features of the present disclosure will become apparent from the following specification. Drawings The drawings are for a better understanding of the present solution and are not to be construed as limiting the present disclosure. Wherein: FIG. 1 is a flow chart of a first embodiment of a flow isolation method according to the present disclosure; Fig. 2 is a first schematic diagram of the HQoS scheduling system and node configuration information of each node therein according to the present disclosure; fig. 3 is a second schematic diagram of the HQoS scheduling system and node configuration information of each node therein according to the present disclosure; FIG. 4 is a flow chart of a second embodiment of a flow isolation method according to the present disclosure; FIG. 5 is a schematic diagram of the structure of a flow isolation device 500 according to the present disclosure; Fig. 6 shows a schematic block diagram of an electronic device 600 that may be used to implement embodiments of the present disclosure. Detailed Description Exemplary embodiments of the present disclosure are described below in conjunction with the accompanying drawings, which include variou