CN117992233A

CN117992233A - Session management method and device, electronic equipment and storage medium

Info

Publication number: CN117992233A
Application number: CN202410217584.1A
Authority: CN
Inventors: 赖诚
Original assignee: Hangzhou DPTech Technologies Co Ltd
Current assignee: Hangzhou DPTech Technologies Co Ltd
Priority date: 2024-02-27
Filing date: 2024-02-27
Publication date: 2024-05-07

Abstract

The application provides a session management method, a session management device, an electronic device and a storage medium, wherein the method is applied to network equipment with a multi-path CPU architecture, and each CPU is preconfigured with a plurality of VCPUs. And respectively distributing at least one VCPU for each node according to the total number of the VCPUs and the total number of the nodes by acquiring the total number of the VCPUs and the total number of the nodes in the network equipment. And determining a target VCPU from the VCPUs allocated to each node, wherein the target VCPU is used for respectively carrying out session aging management on the session in the local memory corresponding to each node, thereby improving the session aging management performance and efficiency of the network equipment of the multi-path CPU architecture.

Description

Session management method and device, electronic equipment and storage medium

Technical Field

The present application relates to the field of computer information processing, and in particular, to a session management method, a session management device, an electronic device, and a storage medium.

Background

Because of the rapid development of multi-core CPUs, each virtual central processing unit VCPU (Virtual Central Processing Unit) needs to access the memory by sharing a north bridge, and as the number of cores increases, the performance of the north bridge in response time becomes a new bottleneck, so that the multi-path/multi-band CPU becomes a new solution.

As shown in fig. 1, each VCPU may access its own corresponding local memory, referred to as local memory access LMA (Local Memory Access), and the VCPUs may also access each other's memory via an internal bus, referred to as remote memory access RMA (Remote Memory Access), so Local Memory Access and Remote Memory Access under the multi-channel/multi-band CPU architecture are collectively referred to as Non-coherent memory access NUMA (Non-Uniform Memory Access).

In the related art, under the multi-path/multi-band CPU architecture, different sessions are allocated to different nodes, and VCPUs responsible for session aging processing are allocated to only one Node, which results in that when VCPUs responsible for session aging processing perform aging processing on sessions on other nodes, inter-Node access to memory occurs, and Remote Memory Access is generated, resulting in performance degradation of session aging management.

Disclosure of Invention

In view of the above, the present application provides a session management method, apparatus, electronic device and storage medium, so as to improve session aging management performance and efficiency of network devices with multi-path CPU architecture.

In a first aspect of the present application, a session management method is provided, applied to a network device with a multi-path CPU architecture, where each CPU is preconfigured with a plurality of VCPUs, including:

Acquiring the total number of VCPUs and the total number of nodes in the network equipment;

Respectively distributing at least one VCPU for each node according to the total number of the VCPUs and the total number of the nodes;

And determining a target VCPU from the VCPUs allocated to each node, wherein the target VCPU is used for performing session aging management on the session in the local memory corresponding to each node.

In an exemplary embodiment of the present application, further comprising:

based on the target VCPU of each node, monitoring a session use chain in a local memory corresponding to the node in real time;

And if the aging time of the session in the session using chain is equal to the preset aging time, performing aging treatment on the session.

In an exemplary embodiment of the present application, the aging the session includes:

marking the index table corresponding to the session as an invalid state;

judging whether a reference count field in the index table is 0;

And if the reference count field in the index table is 0, marking the index table corresponding to the session as a deletion state.

In an exemplary embodiment of the present application, after marking the index table corresponding to the session as a deleted state, the method further includes:

After waiting for a preset delay time, releasing the memory space dynamically applied by the session;

updating the index table corresponding to the session into an idle state, and mounting the session to a session idle chain in a local memory corresponding to the node.

In a second aspect of the present application, there is provided a session management apparatus applied to a network device of a multi-path CPU architecture, each CPU preconfigured with a plurality of VCPUs, including:

An obtaining unit, configured to obtain a total number of VCPUs and a total number of nodes in the network device;

An allocation unit, configured to allocate at least one VCPU to each node according to the VCPU total number and the node total number;

And the processing unit is used for determining a target VCPU from the VCPUs allocated to each node respectively, wherein the target VCPU is used for performing session aging management on the session in the local memory corresponding to each node respectively.

In a third aspect of the application, there is provided an electronic device comprising a processor and a memory storing machine executable instructions executable by the processor for executing the machine executable instructions to implement any of the methods provided in the first aspect.

In a fourth aspect of the application, there is provided a machine-readable storage medium having stored thereon machine-executable instructions which when executed by a processor implement any of the methods provided in the first aspect.

In a fifth aspect of the application, there is provided a computer program product comprising computer programs/instructions which, when executed by a processor, implement the steps of the method provided in the first aspect.

As can be seen from the above technical solution, the present application is applied to a network device with a multi-path CPU architecture, where each CPU is preconfigured with a plurality of VCPUs, and by obtaining the total number of VCPUs and the total number of nodes in the network device, at least one VCPU is allocated to each node according to the total number of VCPUs and the total number of nodes, and a target VCPU is determined from the VCPUs allocated to each node, where the target VCPU is used to perform session aging management on sessions in a local memory corresponding to each node, thereby improving session aging management performance and efficiency of the network device with the multi-path CPU architecture.

Drawings

FIG. 1 is a diagram illustrating memory access under a multi-path CPU architecture according to an embodiment of the present application;

fig. 2 is a flow chart of a session management method according to an embodiment of the present application;

FIG. 3 is a schematic flow chart of VCPU resource initialization according to the embodiment of the present application;

FIG. 4 is a schematic flow diagram of a session aging process according to an embodiment of the present application;

fig. 5 is a schematic structural diagram of a session management device according to an embodiment of the present application;

Fig. 6 is a schematic hardware structure of an electronic device according to an embodiment of the present application.

Detailed Description

Example embodiments will now be described more fully with reference to the accompanying drawings. However, the exemplary embodiments can be embodied in many forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the concept of the example embodiments to those skilled in the art. The same reference numerals in the drawings denote the same or similar parts, and thus a repetitive description thereof will be omitted.

Furthermore, the described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a thorough understanding of embodiments of the application. One skilled in the relevant art will recognize, however, that the application may be practiced without one or more of the specific details, or with other methods, components, devices, steps, etc. In other instances, well-known methods, devices, implementations, or operations are not shown or described in detail to avoid obscuring aspects of the application.

The block diagrams depicted in the figures are merely functional entities and do not necessarily correspond to physically separate entities. That is, the functional entities may be implemented in software, or in one or more hardware modules or integrated circuits, or in different networks and/or processor devices and/or microcontroller devices.

The flow diagrams depicted in the figures are exemplary only, and do not necessarily include all of the elements and operations/steps, nor must they be performed in the order described. For example, some operations/steps may be decomposed, and some operations/steps may be combined or partially combined, so that the order of actual execution may be changed according to actual situations.

Those skilled in the art will appreciate that the drawings are schematic representations of example embodiments and that the modules or flows in the drawings are not necessarily required to practice the application and therefore should not be taken to limit the scope of the application.

First, some technical abbreviations related to the present application will be explained:

DMA: memory direct access (Direct Memory Access), a function provided by some computer bus architectures, enables data to be sent directly from an attached device (e.g., disk drive) to the memory of a computer motherboard.

NUMA: non-uniform memory access (Non Uniform Memory Access), a computer memory design for a multiprocessor, is a memory access time that depends on the memory location of the processor. NUMA architecture classifies memory access latency into two categories: local memory access (Local Memory Access) and remote memory access (Remote Memory Access), wherein a processor accesses its own local memory faster than remote memory (a processor or memory shared between memory and another processor).

LMA: local memory access (Local Memory Access), in a NUMA architecture, a processor sends requests to a memory directly connected thereto (also referred to as a local memory) through its internal bus and reads or writes data directly from the local memory. Because of the close physical distance between the processor and the local memory, the access latency between the processor and the local memory is low.

RMA: remote memory access (Remote Memory Access), in a NUMA architecture, a processor needs to access operations of the remote memory of other processors through its internal bus or network. The remote memory refers to a memory with lower association degree with the current processor, and the access delay between the processor and the remote memory is higher.

VCPU: the virtual central processing unit (Virtual Central Processing Unit) is a logical CPU in a virtualized environment. In virtualization technology, multiple Virtual Machines (VMs) may be created by virtualization software, each of which may be assigned one or more VCPUs. VCPU is a processor emulation of a virtual machine that is responsible for running the operating system and applications within the virtual machine.

Node: a node, an endpoint of a network connection or a point of connection of two (or more) lines. A node may be a processor, controller, or workstation. Nodes vary in their function and may be interconnected together by links to act as control points in a network.

In the initialization stage of VCPU resources, the network equipment divides the VCPU into a control core, an aging core, a data core and the like according to the functional requirements, wherein the aging core is used for carrying out session aging management on sessions in each node of the network equipment. In the related art, in the VCPU resource initialization pipe stage, the network device uniformly distributes the aging core to a certain node based on global management, when a session reaches an aging condition and needs to be aged, if the session needing to be aged is located in the node, the aging core in the node will perform Local Memory Access on the session, and if the session needing to be aged is located in another node except the node, the aging core in the node will perform Remote Memory Access on the session, which causes serious performance loss.

Referring to fig. 2, a flow chart of a session management method according to an embodiment of the present application is provided, and the session management method is applied to a network device with a multi-path CPU architecture, that is, the network device includes a plurality of CPUs, wherein each CPU is preconfigured with a plurality of VCPUs, and the VCPUs are a concept in a virtualization technology, and allow a physical CPU core to be divided into a plurality of logical CPU cores, and each logical core can independently run an operating system and an application program.

As shown in fig. 2, the session management method may include the steps of:

Step 201: and obtaining the total number of VCPUs and the total number of nodes in the network equipment.

Illustratively, in a network device of a multi-path CPU architecture, a "node" generally refers to a physical CPU core, and thus the total number of nodes in the acquiring network device described above may be the number of physical CPU cores in the acquiring network device.

In some embodiments, the network device runs a Linux operating system and supports virtualization, and then a "lscpu" command may be used to obtain detailed information about the CPU of the network device, including the physical CPU core number, thread number, cache size, and so on.

In some embodiments, the network device supports KVM (Kernel-based Virtual Machine) virtualization, and then the "virsh" command may be used to manage virtual machines. For example, the "virsh vcpuinfo < virtual machine name >" command is used to obtain VCPU details of the virtual machine domain, including obtaining the VCPU total number and configuration of the virtual machine.

In some embodiments, the network device supports VMware virtualization, and may use a "VMware-vim-cmd" command to obtain summary information for the virtual machine, including the number of VCPUs for the virtual machine.

It should be noted that the above embodiments are exemplary descriptions of the manner of acquiring the VCPU total number and the node total number in the network device in the present application, and embodiments of the present application are not limited thereto specifically.

In some embodiments, after the total number of nodes in the network device is obtained, a corresponding memory space may be allocated for each node, that is, the local memory mentioned above in the present application. The memory space may be used as a session memory pool for storing session data allocated to the node.

Step 202: and respectively distributing at least one VCPU for each node according to the total number of the VCPUs and the total number of the nodes.

In some embodiments, after the VCPU total and the node total are obtained, each node may be further allocated "VCPU total/node total" VCPUs.

Step 203: and determining a target VCPU from the VCPUs allocated to each node, wherein the target VCPU is used for performing session aging management on the session in the local memory corresponding to each node.

In some embodiments, a preset number of VCPUs may be selected from VCPUs allocated to each node as a target VCPU, where the target VCPU is used to perform session aging management on sessions in a local memory corresponding to each node, that is, the target VCPU is an aging core in the foregoing description of the present application.

In the embodiment of the application, the VCPU is evenly distributed into each node, and each node is respectively allocated with a preset number of ageing cores which are responsible for the session ageing management of the node, so that each node can respectively process the session which needs to be ageing management of the node, the generation of Remote Memory Access of cross-node access memory is avoided, and the session ageing management performance and efficiency of the network equipment of the multi-path CPU architecture are improved.

In some embodiments, after determining the target VCPU from the VCPUs allocated to each node, the session use chain in the local memory corresponding to each node may be monitored in real time based on the target VCPU of each node, and if the aging time of the session in the session use chain is equal to the preset aging time, the session is aged.

For example, session management on each node may be implemented by a chain structure, i.e., a session usage chain. In the session use chain, each session data in the node is used as a linked list element and is linked together through pointers in the linked list.

The elements in the session use chain can be arranged continuously in sequence, so that the access time and the cost of the memory are reduced, more effective management and processing of session data can be realized, and the efficiency of session management is improved.

For example, in the session management, when the aging time of a session is equal to the preset aging time, the session is considered to be expired or invalid, and session aging processing is required for the session.

The setting of the above-mentioned preset aging time needs to be determined according to the session characteristics and the application requirements, such as 20 seconds, 120 seconds, 1200 seconds, etc., which is not particularly limited by the embodiment of the present application. The shorter preset aging time can reduce the storage and management of invalid sessions, and simultaneously can release related resources in time, but can increase the processing burden of the network equipment. The longer preset aging time can reduce the frequency of aging processing and reduce the system overhead of the network equipment, but the invalid session can occupy excessive resources to affect the performance and efficiency of the network equipment system.

In the embodiment of the application, the session aging time in the session use chain in the node is monitored in real time through the allocated aging core of each node, and when the aging time of the session in the node reaches the preset aging time, the session is aged, thereby improving the session aging management performance and efficiency of the network equipment of the multi-path CPU architecture.

In some embodiments, if the aging time of the session in the session usage chain is equal to a preset aging time, aging the session. Wherein aging the session may include the steps of:

S1, marking an index table corresponding to the session as an invalid state.

S2, judging whether a reference count field in the index table is 0.

And S3, if the reference count field in the index table is 0, marking the index table corresponding to the session as a deleting state.

Illustratively, each session in a node has a corresponding index table that can be used to quickly retrieve and manage the session. The fields in the index table at least include a session state and a reference count, and of course, the index table may also include other information such as a session ID, a user ID, a client IP address, a client port, a session type, and the like, which is not particularly limited in the embodiment of the present application.

The session state field is used to indicate the current state of the session corresponding to the index table, such as an active state, an inactive state, a delete state, an idle state, and the like.

In some embodiments, if the aging time of the session in the session usage chain is equal to a preset aging time, it indicates that the session is expired or invalid, and it needs to be aged. At this time, the index table corresponding to the session may be marked as an invalid state, and more specifically, a session state field in the index table corresponding to the session may be marked as an invalid state.

The above reference count field is used to track the number of times the session corresponding to the index table is currently referenced, and the field can record and monitor the use condition of the session, so as to ensure that the session is properly managed and cleaned when appropriate.

In some embodiments, after marking a session state field in an index table corresponding to the session as an invalid state, it is determined whether a reference count field in the index table is "0".

If the reference count field in the index table is not "0," it indicates that the session may be being relied upon by multiple components or services, or that the session may need to be referenced in multiple places, such as in a database, cache, or distributed cache system. In this case, no subsequent processing is performed on the index table, and it is continued to determine whether the reference count field in the index table is "0".

If the reference count field in the index table is "0", this indicates that the session is no longer relied upon by any component or service, the session can be safely destroyed and the associated resources released. In this case, the session state field in the index table corresponding to the session may be marked as a deleted state.

In the embodiment of the application, the session needing aging treatment can be safely destroyed through the session state field and the reference count field in the index table corresponding to the session.

In some embodiments, after marking the index table corresponding to the session as a deleted state, the memory space dynamically applied by the session may be released after waiting for a preset delay time, and the index table corresponding to the session is updated to an idle state, and the session is mounted to a session idle chain in the local memory corresponding to the node.

In the embodiment of the application, the message matched with the session is helped to be processed by setting the preset delay time, so that the session needing to be aged can be destroyed safely, and the network equipment of the multi-path CPU architecture can effectively manage the session and release resources by releasing the memory space of the session and mounting the session to the session idle chain after waiting for the preset delay time.

Referring to fig. 3, a flow chart of VCPU resource initialization according to an embodiment of the present application is shown.

Step 301: obtaining the total number M of VCPUs in the network device

Step 302: acquiring the total number N of nodes in the network equipment

Step 303: M/N VCPUs are respectively allocated to each node

Step 304: selecting a preset number of VCPUs from the VCPUs allocated to each node as aging cores

Referring to fig. 4, a flow chart of a session aging process according to an embodiment of the present application is shown.

After each node is respectively allocated with a preset number of aging cores responsible for session aging management of the node, a session aging process flow shown in fig. 4 may be performed.

Step 401: the session usage chain of the node is monitored at the time of the aging verification in the node.

Step 402: judging whether the aging time of the session in the session using chain reaches the preset aging time.

If the aging time of the session in the session usage chain does not reach the preset aging time, the process goes to step 401.

If the aging time of the session in the session usage chain reaches the preset aging time, step 403 is executed: and judging whether a reference count field in an index table corresponding to the session is 0.

If the reference count field in the index table corresponding to the session is 0, step 404 is executed: and marking the session state field in the index table as a deletion state.

In some embodiments, after marking the session state field in the index table as a deleted state, the memory space dynamically applied by the session may be released after waiting for a preset delay time.

Step 405: updating the session state field in the index table to be in an idle state, and mounting the session to a session idle chain in a local memory corresponding to the node, and jumping to step 401 after the execution of step 405 is completed.

In this embodiment, the aging core on each node only accesses the session use chain on the node, so as to avoid remote memory access and improve session aging management performance and efficiency of the network device of the multi-path CPU architecture.

The foregoing describes the method provided by the present application. The device provided by the application is described below:

fig. 5 is a schematic structural diagram of a session management device according to an embodiment of the present application.

As shown in fig. 5, the apparatus may include:

an obtaining unit 501, configured to obtain a total number of VCPUs and a total number of nodes in the network device;

An allocation unit 502, configured to allocate at least one VCPU to each node according to the VCPU total number and the node total number;

The processing unit 503 is configured to determine a target VCPU from VCPUs allocated to each node, where the target VCPU is configured to perform session aging management on sessions in a local memory corresponding to each node.

In some embodiments, the processing unit 503 is further configured to:

In some embodiments, the processing unit 503 is specifically configured to perform an aging process on the session by:

marking the index table corresponding to the session as an invalid state;

judging whether a reference count field in the index table is 0;

if the reference count field in the index table is 0, marking the index table corresponding to the session as a deletion state;

And/or the number of the groups of groups,

The embodiment of the application also provides a hardware structure. Referring to fig. 6, fig. 6 is a block diagram of an electronic device according to an embodiment of the present application. As shown in fig. 6, the hardware structure may include: a processor and a machine-readable storage medium storing machine-executable instructions executable by the processor; the processor is configured to execute machine-executable instructions to implement the methods disclosed in the above examples of the present application.

Based on the same application concept as the above method, the embodiment of the present application further provides a machine-readable storage medium, where a number of computer instructions are stored, where the computer instructions can implement the method disclosed in the above example of the present application when the computer instructions are executed by a processor.

By way of example, the machine-readable storage medium may be any electronic, magnetic, optical, or other physical storage device that can contain or store information, such as executable instructions, data, and the like. For example, a machine-readable storage medium may be: RAM (Radom Access Memory, random access memory), volatile memory, non-volatile memory, flash memory, a storage drive (e.g., hard drive), a solid state disk, any type of storage disk (e.g., optical disk, dvd, etc.), or a similar storage medium, or a combination thereof.

It is noted that relational terms such as target and object, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.

The foregoing description of the preferred embodiments of the application is not intended to be limiting, but rather to enable any modification, equivalent replacement, improvement or the like to be made within the spirit and principles of the application.

Claims

1. A session management method, applied to a network device with a multi-path CPU architecture, wherein each CPU is preconfigured with a plurality of VCPUs, comprising:

2. The method according to claim 1, characterized in that the method further comprises:

3. The method of claim 2, wherein aging the session comprises:

marking the index table corresponding to the session as an invalid state;

judging whether a reference count field in the index table is 0;

4. A method according to claim 3, wherein after marking the index table corresponding to the session as deleted, the method further comprises:

5. A session management apparatus, applied to a network device of a multi-path CPU architecture, each CPU preconfigured with a plurality of VCPUs, comprising:

6. The apparatus of claim 5, wherein the processing unit is further configured to:

7. The apparatus according to claim 6, wherein the processing unit is configured to perform the aging process on the session by:

marking the index table corresponding to the session as an invalid state;

judging whether a reference count field in the index table is 0;

And/or the number of the groups of groups,

8. An electronic device comprising a processor and a memory, the memory storing machine executable instructions executable by the processor for executing the machine executable instructions to implement the method of any of claims 1-4.

9. A machine-readable storage medium having stored thereon machine-executable instructions which, when executed by a processor, implement the method of any of claims 1-4.

10. A computer program product comprising computer programs/instructions which, when executed by a processor, implement the steps of the method of claim 1.