Method, device, equipment and medium for improving cluster link service carrying capacity

Abstract

This invention relates to the field of communications and proposes a method, apparatus, device, and medium for improving the service carrying capacity of trunking links. The method includes: acquiring theoretical performance data corresponding to trunking links in different link modes under different service scenarios; acquiring actual performance data corresponding to trunking links in different link modes under the current service scenario, and calculating the performance deviation between the actual performance data and the theoretical performance data; in response to the trunking link being in high-performance mode, setting a performance deviation threshold for the high-performance mode and filtering qualified link modes based on the performance deviation threshold; comparing the magnitude of the performance deviation of the qualified link modes and selecting the link mode with the smallest performance deviation for deployment of the trunking link. The method disclosed in this invention enables the trunking to withstand high-pressure front-end services without affecting trunking reliability, thereby improving the availability of the entire system.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of communication, in particular to a method, device, equipment and medium for improving the service carrying capacity of cluster link. BACKGROUND

[0002] At present, array storage, especially high-end multi-control storage, generally uses Roce (RDMA over Converged Ethernet) or IB (InfiniBand) and the like high-speed links or protocols for cache mirroring interconnection in a cluster frame, and generally uses Roce / FC (Fibre Channel) / IB / IP (Internet Protocol) and the like links between cluster frames, and there are numerous choices, but there is no unified optimization according to the service scenario, and in some extremely stressful scenarios, if the cluster link selected by the current service model is not suitable, a cluster communication bottleneck will occur, so that service data cannot be normally delivered, thereby affecting the availability of the whole system. SUMMARY

[0003] Therefore, the present application provides a method, device, equipment and medium for improving the service carrying capacity of cluster link, wherein the method for improving the service carrying capacity of cluster link provided by the present application is based on dynamic management of other modules of the cluster link by a cluster link intelligent optimization management module, in particular, corresponding link intelligent optimization is performed for different service scenarios in a high-performance mode, so that the front-end service under great pressure can be carried in an extraordinary way without affecting the reliability of the cluster, thereby improving the availability of the whole system.

[0004] Based on the above purpose, one aspect of an embodiment of the present application provides a method for improving the service carrying capacity of cluster link, the method comprising the following steps: acquiring theoretical performance data of cluster link corresponding to different link modes under different service scenarios; acquiring actual performance data of cluster link corresponding to different link modes under a current service scenario, and calculating the performance deviation of the actual performance data and the theoretical performance data; in response to the cluster link being in a high-performance mode, setting a performance deviation threshold of the high-performance mode and screening link modes with qualified performance based on the performance deviation threshold of the high-performance mode, comparing the sizes of the performance deviations of the link modes with qualified performance and selecting a link mode with the smallest performance deviation to deploy the cluster link.

[0005] In some embodiments, the operation of setting the performance deviation threshold of the high performance mode and screening the link modes with qualified performance based on the performance deviation threshold of the high performance mode, comparing the sizes of the performance deviations of the link modes with qualified performance, and selecting the link mode with the minimum performance deviation to deploy the cluster link in response to the cluster link being in the high performance mode comprises: setting the performance deviation threshold of the high performance mode and a first duration of testing performance deviation in response to the cluster link being in the high performance mode; screening the link modes with performance deviation less than the performance deviation threshold of the high performance mode for more than the first duration of testing performance deviation to obtain the link modes with qualified performance; comparing the sizes of the performance deviations of the link modes with qualified performance, and selecting the link mode with the minimum performance deviation to deploy the cluster link.

[0006] In some embodiments, the operation of comparing the sizes of the performance deviations of the link modes with qualified performance, and selecting the link mode with the minimum performance deviation to deploy the cluster link comprises: in response to the link mode with the minimum performance deviation being two different link modes, deploying the cluster link through the two different link modes.

[0007] In some embodiments, the operation of comparing the sizes of the performance deviations of the link modes with qualified performance, and selecting the link mode with the minimum performance deviation to deploy the cluster link further comprises: in response to the link mode with the minimum performance deviation being the same single link mode, deploying the cluster link through the same single link mode in redundancy to each other.

[0008] In some embodiments, the method further comprises: in response to the cluster link being in the normal mode, setting the performance deviation threshold of the normal mode, and performing operations corresponding to matching the sizes of the actual performance data and the theoretical data and the sizes of the performance deviations of the different links and the performance deviation threshold of the normal mode.

[0009] In some embodiments, the operation of setting the performance deviation threshold of the normal mode and performing operations corresponding to matching the sizes of the actual performance data and the theoretical data and the sizes of the performance deviations of the different links and the performance deviation threshold of the normal mode in response to the cluster link being in the normal mode comprises: in response to the cluster link being in the normal mode and the actual performance data being greater than the theoretical performance data, setting the performance deviation threshold of the normal mode and a second duration of testing performance deviation, screening the link modes with performance deviation greater than the performance deviation threshold of the normal mode for more than the second duration of testing performance deviation, and updating the corresponding database.

[0010] In some embodiments, the step of setting the performance deviation threshold of the normal mode according to the size of the actual performance data and the theoretical data and the size of the performance deviation of the different link modes and the second duration of the test performance deviation, and screening the link mode whose performance deviation is greater than the performance deviation threshold of the normal mode for more than the second duration of the test performance deviation and determining whether to switch to the high performance mode, further comprises: in response to the cluster link being in the normal mode and the theoretical performance data being greater than the actual performance data, setting the performance deviation threshold of the normal mode and the second duration of the test performance deviation, screening the link mode whose performance deviation is greater than the performance deviation threshold of the normal mode for more than the second duration of the test performance deviation and determining whether to switch to the high performance mode.

[0011] In another aspect of the embodiments of the present application, a device for improving the service carrying capacity of a cluster link is also provided, which comprises: a first module configured to acquire theoretical performance data of the cluster link in different link modes under different service scenarios; a second module configured to acquire actual performance data of the cluster link in different link modes under a current service scenario and calculate the performance deviation of the actual performance data from the theoretical performance data; and a third module configured to, in response to the cluster link being in a high performance mode, set a performance deviation threshold of the high performance mode, screen link modes with qualified performance based on the performance deviation threshold of the high performance mode, compare the size of the performance deviation of the link modes with qualified performance, and select a link mode with the smallest performance deviation to deploy the cluster link.

[0012] In another aspect of the embodiments of the present application, a computer device is also provided, which comprises at least one processor and a memory storing computer instructions executable on the processor, and the instructions are executed by the processor to implement the steps of any of the above methods.

[0013] In another aspect of the embodiments of the present application, a computer readable storage medium is also provided, which stores a computer program executable by a processor to implement the steps of any of the above methods.

[0014] The present application has at least the following beneficial effects: the present application provides a method, device, equipment and medium for improving the service carrying capacity of a cluster link, wherein the method for improving the service carrying capacity of the cluster link dynamically manages different modules of the cluster link, especially intelligently optimizes the link for different service scenarios in the high performance mode, realizes the super-normal carrying of the front-end service under the condition of not affecting the reliability of the cluster, and further improves the availability of the whole system. BRIEF DESCRIPTION OF DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed in the embodiments or prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor based on these drawings.

[0016] Figure 1 A schematic diagram of an embodiment of a method for improving service carrying capacity of a cluster link provided by the present application is shown.

[0017] Figure 2 A schematic diagram of an inter-frame topology of a cluster link provided by the present application is shown.

[0018] Figure 3 A schematic diagram of an embodiment of a device for improving service carrying capacity of a cluster link provided by the present application is shown.

[0019] Figure 4 A schematic diagram of an embodiment of a computer device provided by the present application is shown.

[0020] Figure 5 A schematic diagram of an embodiment of a computer readable storage medium provided by the present application is shown. DETAILED DESCRIPTION

[0021] Embodiments of the present application are described below. However, it should be understood that the disclosed embodiments are merely examples and other embodiments can take various alternative forms.

[0022] In addition, it should be noted that all the expressions of "first" and "second" in the embodiments of the present application are used to distinguish two same name non-same entities or non-same parameters. It can be seen that "first" and "second" are only for the convenience of description and should not be understood as a limitation of the embodiments of the present application. The subsequent embodiments will not be described one by one. The terms "include", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device containing a series of elements not only includes those elements, but also includes elements not explicitly listed or inherent to these processes, methods, articles or devices.

[0023] One or more embodiments of the present application will be described below with reference to the accompanying drawings.

[0024] Based on the above purpose, a first aspect of the embodiments of the present application proposes an embodiment of a method for improving service carrying capacity of a cluster link. Figure 1 A schematic diagram of an embodiment of a method for improving service carrying capacity of a cluster link provided by the present application is shown. As shown in the figure, Figure 1As shown, a method for improving the service carrying capacity of a cluster link according to an embodiment of the present application comprises the following steps:

[0025] S1, obtaining theoretical performance data corresponding to cluster links in different link modes under different service scenarios;

[0026] S2, obtaining actual performance data corresponding to cluster links in different link modes under a current service scenario, and calculating performance deviations of the actual performance data and the theoretical performance data;

[0027] S3, in response to the cluster link being in a high-performance mode, setting a performance deviation threshold of the high-performance mode, screening link modes with qualified performance based on the performance deviation threshold of the high-performance mode, comparing the sizes of the performance deviations of the link modes with qualified performance, and selecting a link mode with the smallest performance deviation to deploy the cluster link.

[0028] To achieve the above purpose, a first aspect of an embodiment of the present application further provides another embodiment of a method for improving system reliability. A method for improving system reliability according to an embodiment of the present application comprises the following steps: through setting a cluster link intelligent optimization management module (CLIO, Cluster link intelligent optimization) as shown, Figure 2 to dynamically manage the self-control cluster link module, the self-control cache module, the self-control CPU module, the opposite-control cluster link module, the opposite-control cache module, and the opposite-control CPU module in the cluster link, as shown. Figure 2 The cluster link intelligent optimization management module is set on a board card, interconnected by high-speed links such as Roce, FC, IB, and IP, mainly uses programmable logic devices such as ARM, and in a physical form of a hot-pluggable interface card, such as an FC card, a network card with Roce function, an IB card, and a general network card. The cache module is located on the board card, generally includes memory and cache in components, and is controlled by the cluster link intelligent optimization management module. The cache modules between the controllers in the frame generally use high-speed links or protocols such as NTB, Roce, or IB for interconnection, such as PCIE multi-control cards, Roce cards, and IB cards. The CPU module is a control module of an array storage, bearing storage system software. In addition to the above modules, the present embodiment further includes an indication module, a wireless module, and a serial port module. The indication module is located on the board card and is directly controlled by the serial port module, and indicates the real-time state of the current CLIO management module to the outside. The wireless module can convert the serial port module signal into a wireless signal such as WIFI, and the outside can interact with the CLIO management module without using physical serial port lines. The serial port module can be used for information interaction between the outside and the CLIO management module, parameter presetting, and the opening of related functions.

[0029] Specifically, for different cluster link modes under different front-end service models, classic data models such as size data blocks, random / sequential, read / write, OLAP (Online Analytical Processing) / OLTP (On-Line Transaction Processing) are embedded into the cluster link intelligent optimization management module. There are two modes between the frames of the cluster link, which are the normal mode and the high-performance mode. In the normal mode and the high-performance mode, the inter-frame link is redundant while the intra-frame cache mirror link is not redundant. In the normal mode, the redundant inter-frame link must select different link modes, such as FC link and Roce link being redundant to each other. In the high-performance mode, the redundant inter-frame link can select different link modes, such as FC link and Roce link being redundant to each other, or FC link and another FC link being redundant to each other. In this mode, the cluster link intelligent optimization management module relies on the embedded database to select the appropriate link mode according to the current business scenario, and changes the physical link mode through simulated hot swapping (in the high-performance mode, there are various interface cards supporting different link modes when the storage is deployed).

[0030] In the normal mode, the cluster link intelligent optimization management module obtains the actual performance data N (including bandwidth data, IOPS, latency, etc.) of the current cluster link mode under different front-end service models, and compares the performance deviation R = |M-N| / M in real time. If R is greater than A (M is less than N) in a link mode for more than a duration T1, the embedded database of the cluster link intelligent optimization management module is updated. If R is greater than A (M is greater than N) in a link mode for more than a duration T1, the cluster link intelligent optimization management module reports that there is a cluster performance bottleneck in this link mode for this business model, and asks whether to switch to the high-performance mode.

[0031] In the high-performance mode, the cluster link intelligent optimization management module relies on the database embedded in it, which can select the appropriate link mode according to the current business scenario, and change the physical link mode through simulated hot swapping. When the cluster link is deployed, a simulated test is performed for a duration T2, and the cluster link intelligent optimization management module compares the performance deviation R = |M-N| / M in real time. When R is less than B for all link modes for more than a duration T3, the R value of the two link modes with the smallest R value is calculated, and the two link modes are used for final business deployment. If there is only one R value at this time, the two single link modes are selected to be redundant to each other for final business deployment.

[0032] A\B\T1\T2\T3\M are system preset parameters, which can be adjusted by the system or a serial port module and the like. Through dynamic management of different modules of the cluster link, especially intelligent optimization of the link for different service scenarios in the high-performance mode, the front-end service under abnormally high pressure can be borne without affecting the reliability of the cluster, thereby improving the availability of the whole system.

[0033] Based on the above purpose, a first aspect of an embodiment of the present application also proposes another embodiment of a method for improving the service bearing capacity of a cluster link. In the embodiment, the steps of the method for improving the service bearing capacity of the cluster link include: dynamically managing the master cluster link module, the master cache module, the master CPU module, the slave cluster link module, the slave cache module and the slave CPU module in the cluster link by setting a cluster link intelligent optimization management module (CLIO, Cluster link intelligent optimization). The cluster link intelligent optimization management module is set on a board card, interconnected by a high-speed link such as Roce, FC, IB and IP, mainly applied to programmable logic devices such as ARM, and in a physical form of a hot-pluggable interface card, such as an FC card, a network card with Roce function, an IB card and a general network card. The cache module is located on the board card and generally includes memory and high-speed cache in the component, controlled by the cluster link intelligent optimization management module, and the cache modules between the controllers in the frame generally interconnected by high-speed links or protocols such as NTB, Roce or IB, such as a PCIE multi-control card, a Roce card and an IB card. The CPU module is a control module of an array storage, bearing storage system software. In addition to the above modules, the embodiment also includes an indication module, a wireless module and a serial port module. The indication module is located on the board card and directly controlled by the serial port module, indicating the real-time state of the CLIO management module to the outside. The wireless module can convert the serial port module signal into a wireless signal such as WIFI, so that the outside can interact with the CLIO management module without physical serial port lines. The serial port module can be used to interact with the CLIO management module, preset parameters and enable related functions.

[0034] Specifically, for different cluster link modes under different front-end service models, classic data models such as size data blocks, random / sequential, read / write, OLAP (Online Analytical Processing) / OLTP (On-Line Transaction Processing) are embedded into the cluster link intelligent optimization management module. There are two modes between the frames of the cluster link, namely the normal mode and the high-performance mode. In the normal mode and the high-performance mode, the inter-frame link is redundant while the intra-frame cache mirror link is not redundant. In the normal mode, the redundant inter-frame link must select different link modes, such as FC link and Roce link being redundant to each other. In the high-performance mode, the redundant inter-frame link can select different link modes, such as FC link and Roce link being redundant to each other, or FC link and another FC link being redundant to each other. In this mode, the cluster link intelligent optimization management module relies on the embedded database to select the appropriate link mode according to the current business scenario, and changes the physical link mode through simulated hot swapping (in the high-performance mode, there are various interface cards supporting different link modes when the storage is deployed).

[0035] In the normal mode, the cluster link intelligent optimization management module obtains the actual performance data N (including bandwidth data, IOPS, latency, etc.) of the current cluster link mode under different front-end service models, and compares the performance deviation R = |M-N| / M in real time. If R is greater than 25% (M is less than N) in a time duration of 20 min or more in a certain link mode, the embedded database of the cluster link intelligent optimization management module is updated. If R is greater than 25% (M is greater than N) in a time duration of 20 min or more in a certain link mode, the cluster link intelligent optimization management module reports that there is a cluster performance bottleneck in this link mode for this business model, and asks whether to switch to the high-performance mode.

[0036] In the high-performance mode, the cluster link intelligent optimization management module relies on the database embedded in its internal, and can select the appropriate link mode according to the current business scenario, and change the physical link mode through simulated hot swapping. When the cluster link is deployed, a simulation test is performed for a duration of 48 h, the performance deviation R = |M-N| / M in various link modes is compared in real time by the cluster link intelligent optimization management module, and the R value when R is less than 10% in a time duration of 15 min or more for all link modes is calculated. The two link modes with the smallest R value are used for final business deployment. If there is only one R value at this time, the two single link modes are selected to be redundant to each other for final business deployment.

[0037] The system preset parameters can be adjusted in the system or in a serial port module or the like in either the normal mode or the high-performance mode. Through dynamic management of different modules of the cluster link, especially intelligent optimization of the link for different service scenarios in the high-performance mode, the front-end service under abnormally high pressure can be borne without affecting the reliability of the cluster, and the availability of the whole system is improved.

[0038] A second aspect of the embodiments of the present application provides a device for improving service bearing capacity of a cluster link. Figure 3 An embodiment of the device for improving service bearing capacity of a cluster link provided by the present application is shown in the schematic diagram. Figure 3 As shown in the figure, the device for improving service bearing capacity of a cluster link provided by the present application comprises: a first module 011 configured to acquire theoretical performance data of the cluster link corresponding to different link modes under different service scenarios; a second module 012 configured to acquire actual performance data of the cluster link corresponding to different link modes under a current service scenario, and calculate performance deviation of the actual performance data and the theoretical performance data; and a third module 013 configured to, in response to the cluster link being in a high-performance mode, set a performance deviation threshold of the high-performance mode, filter link modes with qualified performance based on the performance deviation threshold of the high-performance mode, compare the sizes of the performance deviations of the link modes with qualified performance, and select a link mode with the minimum performance deviation to deploy the cluster link.

[0039] Based on the above purpose, a third aspect of the embodiments of the present application provides a computer device, Figure 4 An embodiment of the computer device provided by the present application is shown in the schematic diagram. Figure 4 As shown in the figure, the embodiment of the computer device provided by the present application comprises the following modules: at least one processor 021; and a memory 022, the memory 022 storing computer instructions 023 executable on the processor 021, the computer instructions 023 being executed by the processor 021 to implement the steps of the method as described above.

[0040] The present application further provides a computer readable storage medium. Figure 5 An embodiment of the computer readable storage medium provided by the present application is shown in the schematic diagram. Figure 5 As shown in the figure, the computer readable storage medium 031 stores a computer program 032 executed by the processor to execute the method as described above.

[0041] Finally, it needs to be explained that all or part of the processes in the above-mentioned embodiment methods can be implemented by a computer program to instruct relevant hardware to complete, and the program of the method of setting system parameters can be stored in a computer readable storage medium. When the program is executed, it can include the processes of the above-mentioned embodiment methods. Among them, the storage medium of the program can be a magnetic disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc. The above-mentioned computer program embodiments can achieve the same or similar effects as the corresponding any method embodiments.

[0042] In addition, the method disclosed in the embodiments of the present application can also be implemented as a computer program executed by a processor, which can be stored in a computer readable storage medium. When the computer program is executed by the processor, the above-mentioned functions defined in the method disclosed in the embodiments of the present application are executed.

[0043] In addition, the above-mentioned method steps and system units can also be implemented by using a controller and a computer readable storage medium for storing a computer program for enabling the controller to implement the above-mentioned steps or unit functions.

[0044] Those skilled in the art will also appreciate that the various illustrative logical blocks, modules, circuits, and algorithm steps described in connection with the disclosure herein can be implemented as electronic hardware, computer software, or combinations of both. To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, and steps have been described generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends on the particular application and design constraints imposed on the overall system. Skilled persons can implement the functions in various ways for each particular application, but such an implementation decision should not be interpreted as causing a departure from the scope of the embodiments disclosed herein.

[0045] In one or more exemplary designs, the functions described can be implemented in hardware, software, firmware, or any combination thereof. If implemented in software, the functions can be stored on or transmitted over as one or more instructions or code on a computer-readable medium. Computer-readable media includes both computer storage media and communication media including any medium that facilitates transfer of a computer program from one place to another. A storage media can be any available media that can be accessed by a general purpose or special purpose computer. By way of example, and not limitation, such computer-readable media can comprise RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to carry or store desired program code means in the form of instructions or data structures and that can be accessed by a general-purpose or special-purpose computer, or a general-purpose or special-purpose processor. Also, any connection is properly termed a computer-readable medium. For example, if the software is transmitted from a website, server, or other remote source using a coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or other wire-based, fiber-based, or wireless technologies, then the coaxial cable, fiber optic cable, twisted pair, DSL, or other wire-based, fiber-based, or wireless technologies are included in the definition of medium. Disk and disc, as used herein, includes compact disc (CD), laser disc, optical disc, digital versatile disc (DVD), floppy disk and blu-ray disc where disks usually reproduce data magnetically, while discs reproduce data optically with lasers. Combinations of the above should also be included within the scope of computer-readable media.

[0046] The foregoing is a summary of the example embodiments disclosed herein, but it should be noted that various changes and modifications can be made without departing from the scope of the embodiments disclosed by the claims. The functions, steps and / or actions of the method claims described herein need not be performed in any particular order. Furthermore, although elements of the embodiments disclosed herein can be described or claimed in individual form, other embodiments can also include a plurality of those elements in combination.

[0047] It should be understood that, as used herein, "a" or "an" can mean one or more things unless context clearly indicates otherwise. It should also be understood that "and / or" as used herein means any and all possible combinations of one or more of the associated listed items.

[0048] The above-mentioned example embodiment numbers of the embodiments disclosed herein are merely for description, and do not represent the advantages or disadvantages of the embodiments.

[0049] Those skilled in the art can understand that all or part of the steps of the above-mentioned embodiments can be completed by hardware, or by program instructing relevant hardware to complete, and the program can be stored in a computer readable storage medium. The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.

[0050] Those skilled in the art shall understand that the above discussion of any embodiment is only exemplary, and is not intended to imply that the scope (including claims) of the embodiments of the present application is limited to these examples; the technical features in the above embodiments or different embodiments can also be combined, and there are many other changes of different aspects of the embodiments of the present application as above. In order to be brief, they are not provided in details. Therefore, any omission, modification, equivalent replacement, improvement, etc. made in the spirit and principle of the embodiments of the present application shall be included in the protection scope of the embodiments of the present application.

Claims

1. A method for improving the traffic carrying capacity of a cluster link, characterized in that, The method comprises: obtaining theoretical performance data of cluster links in different link modes in different service scenarios; obtaining actual performance data of the cluster links in different link modes in a current service scenario, and calculating performance deviations of the actual performance data and the theoretical performance data; in response to the cluster link being in a high-performance mode, setting a performance deviation threshold of the high-performance mode, screening link modes with qualified performance based on the performance deviation threshold of the high-performance mode, comparing sizes of the performance deviations of the link modes with qualified performance, and selecting a link mode with the minimum performance deviation to deploy the cluster link; in a normal mode, selecting different link modes for redundant inter-frame links; in the high-performance mode, the comparing of the sizes of the performance deviations of the link modes with qualified performance and the selecting of the link mode with the minimum performance deviation to deploy the cluster link comprise: in response to the link mode with the minimum performance deviation being two different link modes, deploying the cluster link through the two different link modes; in response to the link mode with the minimum performance deviation being the same link mode, redundantly deploying the cluster link through the same link mode; selecting a suitable link mode according to a current service scenario, and replacing a physical link mode in a simulated hot-swapping manner.

2. The method of claim 1, wherein, The method further comprises: in response to the cluster link being in the normal mode, setting a performance deviation threshold of the normal mode, and performing operations corresponding to matching of sizes of the actual performance data and the theoretical performance data and sizes of the performance deviations of the different links and the performance deviation threshold of the normal mode. The operations corresponding to the matching of the sizes of the actual performance data and the theoretical performance data and the sizes of the performance deviations of the different links and the performance deviation threshold of the normal mode comprise: in response to the cluster link being in the normal mode and the actual performance data being greater than the theoretical performance data, setting a performance deviation threshold of the normal mode and a second duration for testing performance deviations, screening link modes with performance deviations greater than the performance deviation threshold of the normal mode for more than the second duration, and updating a corresponding database.

3. The method of claim 1, wherein, ​ ​ 4. The method of claim 3, wherein, ​ ​ 5. The method of claim 3, wherein, The method further includes: In response to the cluster link being in the normal mode and the theoretical performance data being greater than the actual performance data, setting a performance deviation threshold of the normal mode and a second duration of testing performance deviation, screening a link mode whose performance deviation is greater than the performance deviation threshold of the normal mode for more than the second duration, and determining whether to switch to a high-performance mode.

6. An apparatus for improving cluster link traffic carrying capacity, the apparatus comprising: The method further includes: A first module configured to acquire theoretical performance data of different link modes of the cluster link under different service scenarios; A second module configured to acquire actual performance data of different link modes of the cluster link under a current service scenario, and calculate performance deviation of the actual performance data and the theoretical performance data; A third module configured to, in response to the cluster link being in a high-performance mode, set a performance deviation threshold of the high-performance mode, screen a link mode with qualified performance based on the performance deviation threshold of the high-performance mode, compare performance deviation of the link mode with qualified performance, and select a link mode with minimum performance deviation to deploy the cluster link; In the normal mode, the inter-frame link in redundancy selects different link modes; In the high-performance mode, in response to the link mode with minimum performance deviation being two different link modes, the cluster link is deployed through the two different link modes; In response to the link mode with minimum performance deviation being the same link mode, the cluster link is deployed through the same link mode in redundancy to each other; According to the current service scenario, a suitable link mode is selected, and a physically link mode is replaced in the form of simulated hot plug.

7. A computer device, characterized by The computer program is executed by the processor to implement the steps of the method of any one of claims 1-5. The computer program is executed by the processor to implement the steps of the method of any one of claims 1-5. ​ ​ 8. A computer-readable storage medium storing a computer program, the computer-readable storage medium comprising instructions that, when executed by a computer, cause the computer to perform the method of any one of claims 1 to 7. ​

Images