WO2016188135A1

WO2016188135A1 - Cpu resource configuration method for cluster router and cluster router

Info

Publication number: WO2016188135A1
Application number: PCT/CN2016/071623
Authority: WO
Inventors: 王春艳
Original assignee: 中兴通讯股份有限公司
Priority date: 2015-05-25
Filing date: 2016-01-21
Publication date: 2016-12-01
Also published as: CN106302198A

Abstract

Provided are a CPU resource configuration method for a cluster router and a cluster router. The method comprises: separating functions of various modules in a cluster router system; operating different service modules on different CPUs; and operating different main service modules and standby service modules on different CPUs, wherein the standby service modules are standby modules of the main service modules, and in the case where services of the main service modules cannot operate normally, the standby service modules are converted into the main service modules of the cluster router, thereby solving the problem that CPU resources on the cluster router cannot satisfy service requirements and improving the reliability of the cluster router system.

Description

Cluster router CPU resource configuration method and cluster router

Technical field

The present invention relates to the field of communications, and in particular to a cluster router CPU resource configuration method and a cluster router.

Background technique

In order to cope with the rapid increase of traffic and scale in the network, the router as the core of the network is also accelerating the pace of capacity expansion. Due to its powerful scalability, the cluster router has become the mainstream technology for solving the capacity expansion of the core router. Cluster router, simply That is, multiple routers are interconnected to form a logically integrated cluster router system.

In the related art, the cluster router is composed of a plurality of chassis, each frame has two or more CPUs, and the CPU runs various management modules and service modules. The management module usually performs management functions such as device configuration command delivery, device resource, and status monitoring collection. The service module usually performs routing protocol calculation, label delivery, and specific function processing. The management module can also be used as a special management service. Therefore, for the sake of simplicity, these management modules and service modules of the cluster router are simply referred to as "service modules" in the following description. The stability of these business modules has a great impact on the stability of the cluster system. Each service module of the cluster router runs centrally on one CPU of the cluster system. This requires high performance on the system CPU. When the CPU fails, it affects all services in the system, causing the entire system to crash. Therefore, in order to enhance reliability, some systems propose the concept of a primary CPU and a backup CPU. In addition to the CPU that centrally runs the service module, another backup CPU is added to run backups of all service modules of the primary CPU. This approach enhances reliability. It also runs and backs up all the business modules as a whole. The main CPU completes the operation of all the business modules. When a failure occurs, the entire main CPU's services are switched to the standby CPU. Moreover, since the service modules are still concentrated in one main CPU and one backup CPU, other CPU resources in the cluster router system are not well utilized, and the use of CPU resources is very uneven.

In the related art, the problem that the CPU resources on the cluster router cannot meet the service requirements is solved, and an effective solution has not been proposed yet.

Summary of the invention

The present invention provides a cluster router CPU resource configuration method and a cluster router to solve at least the problem that the CPU resources on the cluster router cannot meet the service requirements.

According to an embodiment of the present invention, a method for configuring a cluster router CPU resource is provided, including:

The cluster router runs different business modules on different CPUs.

In an embodiment of the invention, the cluster router runs different service modules on different CPUs, including:

The cluster router runs different active service modules and standby service modules on different CPUs, where the The service module includes: the primary service module and the standby service module, where the standby service module is a standby module of the primary service module, where the service of the active service module cannot work normally, The standby service module is converted into a main service module of the cluster router.

In an embodiment of the present invention, the cluster router runs a CPU resource management module on the plurality of CPUs, and the resource management module configures one or more of the service modules to configure corresponding CPU resources.

In an embodiment of the invention, in the case that the service module is revoked from the allocated CPU, the resource management module reclaims CPU resources configured for the service module.

In an embodiment of the present invention, the resource management module creates and updates a CPU resource allocation table, where the CPU resource allocation table is used to record an operation of configuring the CPU resource of the service module.

In the embodiment of the present invention, the resource management module configuring corresponding CPU resources for one or more service modules includes:

And the resource management module selects, according to the CPU resource allocation table, a CPU with the lowest load among the plurality of CPUs to run the primary service module or the standby service module.

In an embodiment of the present invention, the resource management module configures one or more service modules with corresponding CPU resources, including at least one of the following:

Selecting, among the plurality of CPUs, a CPU that does not run any of the service modules is allocated to the active service module;

Selecting, by the CPU of the plurality of CPUs that the service module is not running, the standby service module;

If the plurality of CPUs are running the service module, selecting a CPU that only runs the standby service module and runs the standby service module with the least load is allocated to the active service module;

If the plurality of CPUs are running the service module, selecting a CPU that only runs the active service module and runs the least loaded primary module is allocated to the standby service module;

If the plurality of CPUs are all running the service module, and all of the modules are running, the CPU with the least load of the active service module is selected and allocated to the active service module;

If the plurality of CPUs are all running the service module, and both of the main service modules are running, the CPU with the least load of the standby service module is selected to allocate the standby service module.

In an embodiment of the present invention, the resource management module includes an active management module and a standby management module;

The primary management module and the standby management module are run on different CPUs of the plurality of CPUs. When the primary management module is inoperable, the standby management module takes over the primary management module. jobs.

According to another embodiment of the present invention, a cluster router is further provided, including: a service module and a CPU,

Different of the service modules run on different CPUs.

In an embodiment of the present invention, the service module includes: the primary service module and the standby service module:

The different service modules and the standby service module are running on different CPUs, and the standby service module is a standby module of the active service module, and the service of the active service module cannot work normally. In case, the standby service module is converted into a main service module of the cluster router.

In an embodiment of the present invention, the apparatus further includes: a resource management module,

A CPU resource management module is run on the plurality of CPUs, and the resource management module is configured to configure corresponding CPU resources for one or more of the service modules.

The primary management module and the standby management module run on different CPUs. When the primary management module is inoperable, the standby management module takes over the work of the primary management module.

According to the present invention, different primary service modules and standby service modules are run on different CPUs, wherein the standby service module is a standby module of the primary service module, and the service of the primary service module cannot work normally. In this case, the standby service module is converted into the main service module of the cluster router, which solves the problem that the CPU resources on the cluster router cannot meet the service requirements, and improves the reliability of the cluster router system.

DRAWINGS

The drawings described herein are intended to provide a further understanding of the invention, and are intended to be a part of the invention. In the drawing:

1 is a flow chart of a cluster router CPU resource according to an embodiment of the present invention;

2 is a schematic diagram of interaction between a resource management module and each service module according to a preferred embodiment of the present invention;

3 is a schematic flowchart of allocating active and standby CPU resources by a new service module according to a preferred embodiment of the present invention;

4 is a schematic flowchart of a service module switching of a cluster router system according to a preferred embodiment of the present invention;

FIG. 5 is a schematic flow chart of CPU resource recovery of a cluster router system according to a preferred embodiment of the present invention.

detailed description

The invention will be described in detail below with reference to the drawings in conjunction with the embodiments. It should be noted that the embodiments in the present application and the features in the embodiments may be combined with each other without conflict.

It is to be understood that the terms "first", "second" and the like in the specification and claims of the present invention are used to distinguish similar objects, and are not necessarily used to describe a particular order or order.

In this embodiment, a method for configuring a cluster router CPU resource is provided. FIG. 1 is a flowchart of a cluster router CPU resource according to an embodiment of the present invention. As shown in FIG. 1, the process includes the following steps:

In step S102, the cluster router runs different service modules on different CPUs.

Through the above steps, the service modules of the cluster router are no longer concentrated on one CPU of the cluster router, which solves the problem that the CPU resources on the cluster router cannot meet the service requirements, and improves the reliability of the cluster router system.

In this embodiment, the cluster router runs different active service modules and standby service modules on different CPUs, where the service module includes: the primary service module and the standby service module, where the standby service module is The standby module of the active service module is converted into the active service module of the cluster router when the service of the active service module fails to work normally.

In this embodiment, the cluster router may also run a CPU resource management module on the plurality of CPUs, and the resource management module configures corresponding CPU resources for one or more of the service modules. In the case that the service module is revoked from the allocated CPU, the resource management module reclaims the CPU resources configured for the service module. The resource management module creates and updates a CPU resource allocation table for recording the operation of configuring the CPU resource of the service module.

In this embodiment, the resource management module selects, according to the CPU resource allocation table, the CPU with the lowest load or the most suitable resource among the plurality of CPUs to run the primary service module or the standby service module.

In this embodiment, the resource management module configuring the corresponding CPU resource for the one or more service modules may include at least one of the following:

Selecting a CPU that does not run any of the plurality of CPUs to be allocated to the active service module;

Selecting, by the CPU that does not run any service module of the multiple CPUs, the standby service module;

If the plurality of CPUs are running the service module, select a CPU that only runs the active service module and runs the least loaded primary module to allocate the standby service module;

If the plurality of CPUs are running the service module and both of the modules are running, the CPU with the least load of the active service module is selected and allocated to the active service module.

If the plurality of CPUs are running the service module and both of the main service modules are running, selecting a CPU with the least load of the standby service module to allocate the standby service module;

If there is a business requirement, the CPU that is most suitable for the operation of the service module can be selected according to the specific system resources and the characteristics of the service module.

In this embodiment, the resource management module includes an active management module and a standby management module;

The primary management module and the standby management module run on different CPUs of the plurality of CPUs. When the primary management module is inoperable, the standby management module takes over the work of the primary management module.

In another embodiment of the present invention, a cluster router is further provided, including: a service module and a CPU,

Different business modules run on different CPUs.

In this embodiment, the service module includes: the primary service module and the standby service module:

Different primary CPU modules and the standby service module are running on different CPUs, and the standby service module is a standby module of the primary service module. If the service of the active service module fails to work normally, the standby is performed. The service module is converted into the main service module of the cluster router.

In this embodiment, the apparatus further includes: a resource management module,

A CPU resource management module is configured on the plurality of CPUs, and the resource management module is configured to configure corresponding CPU resources for one or more of the service modules.

The following is a detailed description of the preferred embodiments and embodiments.

The preferred embodiment provides a method for rationally allocating system CPU resources, and improves the reliability of the cluster router system.

2 is a schematic diagram of interaction between a resource management module and each service module according to a preferred embodiment of the present invention. The distributed cluster router preferably implemented by the present invention provides a method for balancing and rationally allocating system CPU resources, including a main management module. , standby management module, main service module, standby service module, CPU resource allocation table.

Active management module and standby management module: The management module of the cluster router system responsible for managing CPU resources. The standby management module serves as a standby module of the primary management module, and takes over the work of the primary management module when the primary module fails to work, and switches to the primary management module.

The main service module and the standby service module are: each service module that can be independently operated in the cluster router system, and the standby service module serves as a backup module of the main service module, and takes over the work of the main module when the main module fails to work, and switches Main business module.

CPU resource allocation table: The management module of the cluster router system records the service modules of each CPU running in the system.

The implementation method of the distributed cluster router system in the preferred embodiment includes:

When the cluster router system is started, the system first selects a CPU to run the main management module, which is responsible for allocating and managing each service module. Create and update a CPU resource allocation table.

After the system runs the main management module, select another CPU to run the standby management module as the standby module of the main management module to enhance the reliability of the management module. Update the CPU resource allocation table.

When configuring a service, the system selects the CPU with the lightest CPU load according to the CPU resource allocation table and assigns it to the service as the main CPU of the service module, and selects another CPU as the standby CPU of the service module. Update the CPU resource allocation table.

When a certain service in the system is abnormal, the primary and backup modules are switched, the active module is revoked from the allocated CPU resources, and the standby module is transferred to the active module. The active management module re-allocates the CPU resources for the standby module of the service. , update the CPU resource allocation table.

When deleting a service that has been independently allocated CPU resources, the system reclaims the active and standby CPU resources allocated to it and updates the CPU resource allocation table.

After the cluster router is started, when no service is configured, the CPU module is not actively allocated to related service modules. Further, when the system selects a CPU for the service module, the CPU resource that is not running any service module is preferentially allocated to the service module; if there is no CPU resource of any service module in the system, the standby module that runs only the standby module and runs is selected. The least amount of CPU resources are allocated to the new service loading main module, and the CPU module that runs only the main module and runs the running main module is allocated to the new service loading standby module; if the system does not only run the standby module For the CPU resources of the module, select the CPU resource with the least number of active modules to allocate to the new service load master module; if there is no CPU resource running only the active module in the system, select the CPU with the least number of spare modules. The resource is allocated to the new service to load the standby module.

The cluster router system of the preferred embodiment of the present invention balances the load of each CPU in the cluster router system, so that the CPU resources in the cluster router system are effectively utilized, and each service module allocated by the system has a backup module, and The main module is allocated on the same CPU. When the main module is abnormal, the standby module can be autonomously transferred to the main module. Since each service module in the system is independent, when a certain service module in the system is abnormal, When the switchover of the active and standby modules is performed, the other independent service modules are not affected, and the service burden of the active CPU is also reduced, thereby improving the overall reliability of the system.

The hardware module includes: Take the 2+2 cluster router system as an example. There are four chassis in the system. Each chassis has two master CPUs. There are 8 CPU resources available in the system. The two CPUs of the center frame 1 are C1 and C2, the two CPUs of the center frame 2 are C3 and C4, the two CPUs of the line card frame 1 are L1 and L2, and the two CPUs of the line card frame 2 are L3 and L4.

The software module part includes: an independently assignable service module in the cluster system, and assumes that the system has five service modules that can be independently allocated, respectively, A, B, C, D, and E;

The processing steps of the CPU resource allocation process part are as follows:

After the cluster router system is started, the active and standby CPUs of the management module are selected independently. Assume that C1 and C3 are selected respectively. The system runs normally without any service. The CPUs in the CPU resource allocation table maintained by the management module are empty:

FIG. 3 is a schematic diagram of a process of allocating active and standby CPU resources by a new service module according to a preferred embodiment of the present invention. As shown in FIG. 3, the system is configured with the A service: according to the process in FIG. 3, the system query management module is first maintained. If there is any CPU resource that does not load any service module, select one of them, such as C2, and assign it to the A service to load the main service module. If the CPU resource table is still queried, the CPU resources of any service module are still not loaded. One of them, for example, C4 is assigned to the A service to load the standby service module; at the same time, the allocation information is added to the CPU resource allocation table;

The system is configured with the B service. According to the rules of the preceding steps, the L1 is allocated to the B service to load the main service module. The L3 is loaded with the standby service module for the B service, and the allocation information is added to the CPU resource allocation table.

The system is configured with the C service: according to the rules of the above steps, the L2 is allocated to the C service to load the main service module, the L4 is allocated to the C service to load the standby service module, and the allocation information is added to the CPU resource allocation table;

The D service is configured in the system. According to the rules in Figure 3, the CPU resources of any service module are not loaded in the system. Then, only the CPU resources of the standby module are allocated, such as C3, and the D service is loaded with the main service module. The C2 of the standby service module loads the standby service module for the D service, and adds the allocation information to the CPU resource allocation table;

The E service is configured in the system. According to the rules in Figure 3 above, the C4 is allocated to the E service to load the main service module, the L1 is allocated to the E service to load the standby service module, and the allocation information is added to the CPU resource allocation table.

4 is a schematic flowchart of a service module switching of a cluster router system according to a preferred embodiment of the present invention. As shown in FIG. 4, when a service main module in the system is abnormal, such as a C service module, the C on the L2 CPU is revoked. The service main module converts the C service standby module on the L4 CPU into the C service main module, and updates the CPU resource table of the management module. At this time, the L2 of the CPU resource table in the management module is not loaded with any service. According to the rule of FIG. 4, the L2 is allocated to the C service to load the standby module, and the CPU resource allocation table of the management module is updated;

When a service main module in the system is abnormal, such as the A service module, the A service main module on the C2 CPU is revoked, and the A service standby module on the C4 CPU is converted into the A service main module, and the management is updated. The CPU resource allocation table of the module. At this time, there is no CPU resource in the CPU resource table of the management module that does not load any service. According to the rule of FIG. 4, the L4 that loads only one main service module can be allocated to load the standby module for the A service. At the same time, update the CPU resource allocation table of the management module.

5 is a schematic flowchart of CPU resource recovery of a cluster router system according to a preferred embodiment of the present invention. As shown in FIG. 5, when a cluster system deletes a certain service, such as service E, the management module reclaims the allocation to E according to the process of FIG. The CPU resource of the service, and the CPU resource allocation table of the management module is updated.

The preferred embodiment of the present invention mainly utilizes each CPU resource in the cluster router to distribute the functions of each module in the system to the available CPU resources in the system, thereby reducing the system performance requirements of the system and reducing the CPU performance. The burden on the main CPU increases the reliability of the system and makes full use of the existing resources in the system.

Through the description of the above embodiments, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, by hardware, but in many cases, the former is A better implementation. Based on such understanding, the technical solution of the present invention, which is essential or contributes to the prior art, may be embodied in the form of a software product stored in a storage medium (such as ROM/RAM, disk, The optical disc includes a number of instructions for causing a terminal device (which may be a mobile phone, a computer, a server, or a network device, etc.) to perform the method of various embodiments of the present invention.

Embodiments of the present invention also provide a storage medium. Optionally, in this embodiment, the foregoing storage medium may be configured to store program code for performing the method steps of the above embodiment:

Optionally, in this embodiment, the foregoing storage medium may include, but is not limited to, a U disk, a read only memory (ROM, Read-Only Memory), Random Access Memory (RAM), removable hard disk, disk or optical disk, etc., which can store program code.

Optionally, in this embodiment, the processor performs the method steps of the foregoing embodiments according to the stored program code in the storage medium.

It will be apparent to those skilled in the art that the various modules or steps of the present invention described above can be implemented by a general-purpose computing device that can be centralized on a single computing device or distributed across a network of multiple computing devices. Alternatively, they may be implemented by program code executable by the computing device such that they may be stored in the storage device by the computing device and, in some cases, may be different from the order herein. The steps shown or described are performed, or they are separately fabricated into individual integrated circuit modules, or a plurality of modules or steps thereof are fabricated as a single integrated circuit module. Thus, the invention is not limited to any specific combination of hardware and software.

The above description is only for the preferred embodiment of the present invention, and is not intended to limit the present invention. For those skilled in the art, the present invention may have various changes and changes, such as a CPU resource scheduling algorithm, which may be based on the hardware of the system. And software conditions, choose a different and more suitable method. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and scope of the present invention are intended to be included within the scope of the present invention.

Industrial applicability

According to the foregoing technical solution provided by the embodiment of the present invention, different primary service modules and standby service modules are run on different CPUs, where the standby service module is a standby module of the primary service module, and the primary service module is used in the primary service module. When the service is not working properly, the standby service module is converted into the main service module of the cluster router, which solves the problem that the CPU resources on the cluster router cannot meet the service requirements, and improves the reliability of the cluster router system.

Claims

A method for configuring a cluster router CPU resource includes:

The cluster router runs different business modules on different CPUs.
The method of claim 1 wherein the cluster router running different service modules on different CPUs comprises:

The cluster router runs different active service modules and standby service modules on different CPUs, where the service module includes: the primary service module and the standby service module, where the standby service module is The standby module of the active service module is converted into the active service module of the cluster router if the service of the active service module fails to work normally.
The method of claim 1 wherein the method further comprises:

The cluster router runs a CPU resource management module on the plurality of CPUs, and the resource management module configures corresponding CPU resources to one or more of the service modules.
The method of claim 3, wherein the method further comprises:

In a case where the service module is revoked from the allocated CPU, the resource management module reclaims CPU resources configured for the service module.
The method of claim 3, further comprising:

The resource management module creates and updates a CPU resource allocation table, where the CPU resource allocation table is used to record the running of the CPU resource of the service module.
The method of claim 5, wherein the resource management module configuring the corresponding CPU resources for the one or more service modules comprises:

And the resource management module selects, according to the CPU resource allocation table, a CPU with the lowest load among the plurality of CPUs to run the primary service module or the standby service module.
The method according to claim 5, wherein the resource management module configures one or more service modules with corresponding CPU resources, including at least one of the following:

Selecting, among the plurality of CPUs, a CPU that does not run any of the service modules is allocated to the active service module;

Selecting, by the CPU of the plurality of CPUs that the service module is not running, the standby service module;

If the plurality of CPUs are running the service module, selecting a CPU that only runs the standby service module and runs the standby service module with the least load is allocated to the active service module;

If the plurality of CPUs are running the service module, selecting a CPU that only runs the active service module and runs the least loaded primary module is allocated to the standby service module;

If the plurality of CPUs are all running the service module, and both of the modules are running, select the running master. Allocating the primary service module to the CPU with the least load of the service module;

If the plurality of CPUs are all running the service module, and both of the main service modules are running, the CPU with the least load of the standby service module is selected to allocate the standby service module.
The method according to any one of claims 3 to 7, wherein the resource management module comprises an active management module and a standby management module;

The primary management module and the standby management module are run on different CPUs of the plurality of CPUs. When the primary management module is inoperable, the standby management module takes over the primary management module. jobs.
A cluster router includes: a service module and a CPU,

Different of the service modules run on different CPUs.
The apparatus according to claim 9, wherein the service module comprises: the primary service module, the standby service module:

The different service modules and the standby service module are running on different CPUs, and the standby service module is a standby module of the active service module, and the service of the active service module cannot work normally. In case, the standby service module is converted into a main service module of the cluster router.
The apparatus of claim 9, wherein the apparatus further comprises: a resource management module,

A CPU resource management module is run on the plurality of CPUs, and the resource management module is configured to configure corresponding CPU resources for one or more of the service modules.
The apparatus according to claim 11, wherein the resource management module comprises an active management module and a standby management module;

The primary management module and the standby management module run on different CPUs. When the primary management module is inoperable, the standby management module takes over the work of the primary management module.