CN118740743A - Data transmission method and system - Google Patents

Abstract

The embodiment of the present application provides a data transmission method and system for reducing the probability of an application server retransmitting a data read and write request. The storage system includes a storage server and an application server, the storage server includes multiple controllers, the application server is connected to at least two of the multiple controllers, and the method includes: when the data processing efficiency of the first controller of the at least two controllers is lower than that of any controller of the at least two controllers except the first controller, the storage server sends a first message to the application server; the application server switches the data to be transmitted through the first link to the second link, and the second link is a link between the second controller of the at least two controllers and the application server.

Description

Data transmission method and system

Technical Field

The present application relates to the field of cloud storage technology, and in particular to a data transmission method and system.

Background Art

Network attached storage (NAS) is a data storage method connected to a computer network, and is often used in the field of file storage. The device used to provide data storage services can be called a storage server or NAS device.

The storage server may include multiple controllers, and the application server can access the controllers through the links between the controllers to read data from the storage server or store data in the storage server. However, when the ability of a controller in the storage server to process data is reduced due to phenomena such as hard disk aging, the efficiency of the controller in processing data will be lower than that of other controllers in the storage server. Then, when the controller receives a new data read and write request, it will discard the new data read and write request due to insufficient data processing capacity, and request the application server to resend the new data read and write request. This will cause the link between the controller and the application server to retransmit data read and write requests more times, which may cause link congestion.

Summary of the invention

The embodiments of the present application provide a data transmission method and system for reducing the probability of an application server retransmitting a data read and write request.

In a first aspect, a data transmission method is provided, which can be implemented by a storage system, the storage system comprising a storage server and an application server, the storage server comprising multiple controllers, the application server being connected to at least two of the multiple controllers, the method comprising: when a data processing efficiency of a first controller of the at least two controllers is lower than that of any other controller of the at least two controllers except the first controller, the storage server sending first information to the application server; wherein the first information is used to instruct the application server to switch a link, the first information comprising an identifier of a first link, the first link being a link between the first controller and the application server; the application server switches the data to be transmitted through the first link to a second link, the second link being a link between the second controller of the at least two controllers and the application server.

In the embodiment of the present application, if the data processing efficiency of a controller is lower than that of other controllers, the storage server can instruct the application server to migrate the data to be transmitted through a link (e.g., the first link) between the controller and the application server to the link of other controllers for transmission. In this way, the amount of data to be processed by the controller with lower data processing efficiency can be reduced, the number of times the application server retransmits data read and write requests can be reduced, the probability of congestion can be reduced, and the performance of the storage system can be improved.

In a possible implementation manner, the data processing efficiency of the first controller is lower than that of any of the controllers, including one or more of the following: the difference between the amount of data processed by any of the controllers and the amount of data processed by the first controller in one cycle is greater than a first threshold, and the amount of data processed by any of the controllers is greater than the amount of data processed by the first controller; the difference between the input/output operations per second (IOPS) of any of the controllers and the IOPS of the first controller is greater than a second threshold, and the IOPS of any of the controllers is greater than the IOPS of the first controller; the first controller has a central processing unit (CPU) a first processing unit (CPU) core isolation event; a difference between a first duration taken by the first controller to process a first amount of data and a second duration taken by any one of the controllers to process the first amount of data is greater than a third threshold, and the first duration is greater than the second duration; a difference between the CPU occupancy of the first controller and the CPU occupancy of any one of the controllers is greater than a fourth threshold, and the CPU occupancy of the first controller is greater than the CPU occupancy of any one of the controllers; a difference in IOPS of the first controller within two adjacent seconds is greater than a fifth threshold; or, within a preset duration, the number of times the first controller sends second information to the application server is greater than a sixth threshold, and the second information is used to request the application server to resend a data read and write request.

In the above technical solution, if the data processing volume of the first controller in a cycle is less than that of any other controller in the multiple controllers, or if the time taken by the first controller to process a certain amount of data is longer than that of any controller, or the CPU occupancy rate of the first controller is greater than that of any controller, it indicates that the speed of the first controller to process data is slowed down and the data processing capacity is reduced. If a core isolation event occurs in the first controller, it indicates that at least one core used for processing data in the first controller fails and cannot continue to process data, so that the first controller has fewer cores for data processing and the data processing capacity is reduced. If the difference in IOPS of the first controller within two consecutive seconds is greater than the fifth threshold, it indicates that the fluctuation of the data processed by the first controller is large and the performance is unstable. If the data processing efficiency of the first controller is reduced, the first controller may discard the new data due to insufficient data processing capacity when receiving new data, and the first controller may also request the application server to resend the new data, for example, the second information sent by the first controller to the application server is used to request the application server to resend the new data. Then, if the number of times the first controller sends the second information to the application server within the preset time is greater than the sixth threshold, it indicates that the data discarded by the first controller is large, that is, the data processing capacity of the first controller is low.

In a possible implementation manner, the method further includes: the storage server setting the state of the first link to a low priority state.

The priority state can be used by the controller to select a link, for example, the controller preferentially selects a link with a high priority. Therefore, the storage server sets the first link to a low priority state, which can reduce the probability that the controller selects the first link to transmit data, thereby helping to improve data transmission efficiency.

In a possible implementation, the method further includes: the storage server sending third information to the application server, the third information being used to instruct the application server to update the state of the first link to the low priority state; and the application server setting the state of the first link to the low priority state.

In the above technical solution, the storage server sends the link status to the application server and instructs the application server to perform status synchronization, which helps to reduce the probability of the application server transmitting data through a low-priority link.

In a possible implementation manner, the method further includes: the application server determining that the link with the smallest link load is the second link; or, the first information is further used to indicate the second link.

In the above technical solution, the storage server performs load balancing on the links corresponding to all controllers to determine the second link, which can make the load balancing degree of the link and/or the load balancing degree of the controller higher. The load balancing degree of the link is used to indicate the difference between the amount of data transmitted through different links; the load balancing degree of the controller is used to indicate the difference in the amount of data processed by different controllers. If the difference is small, it indicates that the load balancing degree is high; if the difference is large, it indicates that the load balancing degree is low. The application server determines the second link by itself. Since the link corresponding to the application server can be less than the link corresponding to the storage server, the number of links that need to be balanced in the load balancing process can be reduced, which helps to improve the efficiency of determining the second link.

In a possible implementation, the first information is also used to indicate the second link, and the method further includes: the storage server obtains load information of at least one link between the application server and a controller other than the first controller among the at least two controllers, and determines that the link with the smallest load among the at least one link is the second link; or, the storage server obtains load information of other controllers other than the first controller among the at least two controllers, determines that the controller with the smallest load is the second controller, and determines that the link with the smallest load among the links between the second controller and the application server is the second link.

In the above technical solution, the storage server selects the link with the smallest load from the links corresponding to all controllers, which can make the load balance between the links higher; the storage server first determines the controller with the smallest load, and selects the link with the smallest load from the controller with the smallest load, which can make the load balance between the controllers higher.

In a second aspect, a data transmission method is provided, which can be applied to a storage server in a storage system, the storage system comprising the storage server and an application server, the storage server comprising multiple controllers, the application server being connected to at least two of the multiple controllers, the method comprising: when a data processing efficiency of a first controller of the at least two controllers is lower than that of any other controller of the at least two controllers except the first controller, sending first information to the application server; wherein the first information is used to instruct the application server to switch a link, the first information comprises an identifier of a first link, and the first link is a link between the first controller and the application server.

In a possible implementation manner, the data processing efficiency of the first controller is lower than that of any of the controllers, including one or more of the following: the difference between the amount of data processed by any of the controllers and the amount of data processed by the first controller within a cycle is greater than a first threshold, and the amount of data processed by any of the controllers is greater than the amount of data processed by the first controller; the difference between the IOPS of any of the controllers and the IOPS of the first controller is greater than a second threshold, and the IOPS of any of the controllers is greater than the IOPS of the first controller; a CPU core isolation event occurs in the first controller; the difference between a first duration used by the first controller to process a first amount of data and a second duration used by any of the controllers to process the first amount of data is greater than a third threshold, and the first duration is greater than the second duration; the difference between the CPU occupancy rate of the first controller and the CPU occupancy rate of any of the controllers is greater than a fourth threshold, and the CPU occupancy rate of the first controller is greater than the CPU occupancy rate of any of the controllers; the difference between the IOPS of the first controller within two adjacent seconds is greater than a fifth threshold; or, within a preset duration, the number of times the first controller sends the second information to the application server is greater than a sixth threshold, and the second information is used to request the application server to resend the data read and write request.

In a possible implementation manner, the method further includes: setting the state of the first link to a low priority state.

In a possible implementation manner, the method further includes: sending third information to the application server, where the third information is used to instruct the application server to update the state of the first link to the low priority state.

In a possible implementation manner, the first information is further used to indicate a second link, where the second link is a link between a second controller of the at least two controllers and the application server.

Obtain load information of at least one link between the application server and a controller other than the first controller among the at least two controllers, and determine that the link with the smallest load among the at least one link is the second link; or, obtain load information of other controllers other than the first controller among the at least two controllers, determine that the controller with the smallest load among them is the second controller, and determine that the link with the smallest load among the links between the second controller and the application service is the second link.

According to a third aspect, a data transmission method is provided, which can be applied to an application server in a storage system, wherein the storage system comprises a storage server and the application server, the storage server comprises multiple controllers, and the application server is connected to at least two of the multiple controllers, the method comprising: receiving first information from the storage server; wherein the first information is sent by the storage server when the data processing efficiency of the first controller among the at least two controllers is lower than that of any other controller among the at least two controllers except the first controller, the first information is used to instruct the application server to switch links, the first information comprises an identifier of a first link, the first link is a link between the first controller and the application server; and the data to be transmitted through the first link is switched to the second link, the second link is a link between the second controller among the at least two controllers and the application server.

In a possible implementation manner, the data processing efficiency of the first controller is lower than that of any of the controllers, including one or more of the following: the difference between the amount of data processed by any of the controllers and the amount of data processed by the first controller within a cycle is greater than a first threshold, and the amount of data processed by any of the controllers is greater than the amount of data processed by the first controller; the difference between the IOPS of any of the controllers and the IOPS of the first controller is greater than a second threshold, and the IOPS of any of the controllers is greater than the IOPS of the first controller; a CPU core isolation event occurs in the first controller; the difference between a first duration used by the first controller to process a first amount of data and a second duration used by any of the controllers to process the first amount of data is greater than a third threshold, and the first duration is greater than the second duration; the difference between the CPU occupancy rate of the first controller and the CPU occupancy rate of any of the controllers is greater than a fourth threshold, and the CPU occupancy rate of the first controller is greater than the CPU occupancy rate of any of the controllers; the difference between the IOPS of the first controller within two adjacent seconds is greater than a fifth threshold; or, within a preset duration, the number of times the first controller sends the second information to the application server is greater than a sixth threshold, and the second information is used to request the application server to resend the data read and write request.

In a possible implementation, the method further includes: receiving third information from the storage server, the third information being used to indicate that the state of the first link is updated to a low priority state; and setting the state of the first link to the low priority state.

In a possible implementation manner, the first information is also used to indicate the second link.

In a possible implementation, the method further includes: determining that the link with the smallest link load is the second link.

In a fourth aspect, a data transmission device is provided, the data transmission device is used to implement the second aspect or any one of the methods in the second aspect, and the data transmission device includes corresponding functional modules, which are respectively used to implement the steps in the above method. The functions can be implemented by hardware, or by hardware executing corresponding software. The hardware or software includes one or more modules corresponding to the above functions.

In a possible implementation, the data transmission device may be arranged in a storage server in a storage system, or the data transmission device is the storage server, the storage system includes the storage server and an application server, the storage server includes multiple controllers, the application server is connected to at least two of the multiple controllers, and the device includes: a processing module and a communication module, the processing module is used to control the communication module to send first information to the application server when the data processing efficiency of the first controller among the at least two controllers is lower than any other controller among the at least two controllers except the first controller; wherein the first information is used to instruct the application server to switch the link, the first information includes an identifier of the first link, and the first link is the link between the first controller and the application server.

In a possible implementation manner, the data processing efficiency of the first controller is lower than that of any of the controllers, including one or more of the following: the difference between the amount of data processed by any of the controllers and the amount of data processed by the first controller within a cycle is greater than a first threshold, and the amount of data processed by any of the controllers is greater than the amount of data processed by the first controller; the difference between the IOPS of any of the controllers and the IOPS of the first controller is greater than a second threshold, and the IOPS of any of the controllers is greater than the IOPS of the first controller; a CPU core isolation event occurs in the first controller; the difference between a first duration used by the first controller to process a first amount of data and a second duration used by any of the controllers to process the first amount of data is greater than a third threshold, and the first duration is greater than the second duration; the difference between the CPU occupancy rate of the first controller and the CPU occupancy rate of any of the controllers is greater than a fourth threshold, and the CPU occupancy rate of the first controller is greater than the CPU occupancy rate of any of the controllers; the difference between the IOPS of the first controller within two adjacent seconds is greater than a fifth threshold; or, within a preset duration, the number of times the first controller sends the second information to the application server is greater than a sixth threshold, and the second information is used to request the application server to resend the data read and write request.

In a possible implementation manner, the processing module is further configured to: set the state of the first link to a low priority state.

In a possible implementation manner, the communication module is further used to: send third information to the application server, where the third information is used to instruct the application server to update the state of the first link to the low priority state.

In a possible implementation manner, the first information is further used to indicate a second link, where the second link is a link between a second controller of the at least two controllers and the application server.

In a possible implementation, the processing module is further used to: obtain load information of at least one link between the application server and a controller other than the first controller among the at least two controllers, and determine that the link with the smallest load among the at least one link is the second link; or, obtain load information of other controllers other than the first controller among the at least two controllers, determine that the controller with the smallest load is the second controller, and determine that the link with the smallest load among the links between the second controller and the application service is the second link.

In a fifth aspect, another data transmission device is provided, which is used to implement the third aspect or any one of the methods in the third aspect, and the data transmission device includes corresponding functional modules, which are respectively used to implement the steps in the above method. The functions can be implemented by hardware, or by hardware executing corresponding software. The hardware or software includes one or more modules corresponding to the above functions.

In a possible implementation, the data transmission device may be arranged in an application server in a storage system, or the data transmission device is the application server, the storage system includes a storage server and the application server, the storage server includes multiple controllers, the application server is connected to at least two of the multiple controllers, the device includes a communication module and a processing module, the communication module is used to receive first information from the storage server; wherein, the first information is sent by the storage server when the data processing efficiency of the first controller among the at least two controllers is lower than that of any other controller among the at least two controllers except the first controller, the first information is used to instruct the application server to switch the link, the first information includes an identifier of the first link, the first link is the link between the first controller and the application server; the processing module is used to switch the data to be transmitted through the first link to a second link, the second link is the link between the second controller among the at least two controllers and the application server.

In a possible implementation manner, the data processing efficiency of the first controller is lower than that of any of the controllers, including one or more of the following: the difference between the amount of data processed by any of the controllers and the amount of data processed by the first controller within a cycle is greater than a first threshold, and the amount of data processed by any of the controllers is greater than the amount of data processed by the first controller; the difference between the IOPS of any of the controllers and the IOPS of the first controller is greater than a second threshold, and the IOPS of any of the controllers is greater than the IOPS of the first controller; a CPU core isolation event occurs in the first controller; the difference between a first duration used by the first controller to process a first amount of data and a second duration used by any of the controllers to process the first amount of data is greater than a third threshold, and the first duration is greater than the second duration; the difference between the CPU occupancy rate of the first controller and the CPU occupancy rate of any of the controllers is greater than a fourth threshold, and the CPU occupancy rate of the first controller is greater than the CPU occupancy rate of any of the controllers; the difference between the IOPS of the first controller within two adjacent seconds is greater than a fifth threshold; or, within a preset duration, the number of times the first controller sends the second information to the application server is greater than a sixth threshold, and the second information is used to request the application server to resend the data read and write request.

In a possible implementation, the communication module is further used to: receive third information from the storage server, the third information being used to indicate that the state of the first link is updated to a low priority state; and the processing module is further used to: set the state of the first link to the low priority state.

In a possible implementation manner, the first information is also used to indicate the second link.

In a possible implementation manner, the processing module is further configured to: determine that the link with the smallest link load is the second link.

In a sixth aspect, a storage system is provided, which includes a storage server and an application server, the storage server includes multiple controllers, and the application server is connected to at least two of the multiple controllers, wherein the storage server is used to send first information to the application server when the data processing efficiency of a first controller among the at least two controllers is lower than that of any controller among the at least two controllers except the first controller; wherein the first information is used to instruct the application server to switch a link, the first information includes an identifier of a first link, the first link being a link between the first controller and the application server; the application server is used to switch data to be transmitted through the first link to a second link, the second link being a link between the second controller among the at least two controllers and the application server.

In a possible implementation manner, the data processing efficiency of the first controller is lower than that of any of the controllers, including one or more of the following: the difference between the amount of data processed by any of the controllers and the amount of data processed by the first controller in one cycle is greater than a first threshold, and the amount of data processed by any of the controllers is greater than the amount of data processed by the first controller; the difference between the number of reads and writes per second (IOPS) of any of the controllers and the IOPS of the first controller is greater than a second threshold, and the IOPS of any of the controllers is greater than the IOPS of the first controller; a CPU core isolation event occurs in the first controller; the difference between a first duration used by the first controller to process a first amount of data and a second duration used by any of the controllers to process the first amount of data is greater than a third threshold, and the first duration is greater than the second duration; the difference between the CPU occupancy rate of the first controller and the CPU occupancy rate of any of the controllers is greater than a fourth threshold, and the CPU occupancy rate of the first controller is greater than the CPU occupancy rate of any of the controllers; the difference between the IOPS of the first controller within two adjacent seconds is greater than a fifth threshold; or, within a preset duration, the number of times the first controller sends the second information to the application server is greater than a sixth threshold, and the second information is used to request the application server to resend the data read and write request.

In a possible implementation manner, the storage server is further configured to: set the state of the first link to a low priority state.

In a possible implementation, the storage server is further used to: send third information to the application server, wherein the third information is used to instruct the application server to update the state of the first link to the low priority state; the application server is further used to: set the state of the first link to the low priority state.

In a possible implementation manner, the application server is used to determine that the link with the smallest link load is the second link; or the first information is also used to indicate the second link.

In a possible implementation, the first information is also used to indicate the second link, and the storage server is further used to: obtain load information of at least one link between the application server and a controller other than the first controller among the at least two controllers, and determine that the link with the smallest load among the at least one link is the second link; or, obtain load information of other controllers other than the first controller among the at least two controllers, determine that the controller with the smallest load is the second controller, and determine that the link with the smallest load among the links between the second controller and the application service is the second link.

In a seventh aspect, a computing device cluster is provided, which includes at least one computing device, each computing device including a processor and a memory, wherein the processor is used to execute instructions stored in the memory of the at least one computing device, so that the computing device cluster executes the method provided in the second or third aspect above.

In an eighth aspect, a computer-readable storage medium is provided, wherein the computer-readable storage medium is used to store a computer program. When the computer program is run on a computer, the computer executes the method provided in the second or third aspect above.

In a ninth aspect, a computer program product is provided, comprising a computer program, which, when executed on a computer, enables the computer to execute the method provided in the second or third aspect above.

In a tenth aspect, a chip system is provided, comprising a processor and an interface, wherein the processor is used to call and execute instructions from the interface so that the chip system implements the method provided in the second or third aspect above.

For the beneficial effects of the second to tenth aspects mentioned above, please refer to the beneficial effects of the first aspect and will not be repeated here.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a schematic diagram of the structure of a storage system provided in an embodiment of the present application;

FIG2A is a schematic diagram of the structure of another storage system provided in an embodiment of the present application;

FIG2B is a schematic diagram of the structure of another storage system provided in an embodiment of the present application;

FIG3 is a schematic diagram of a flow chart of a data transmission method provided in an embodiment of the present application;

FIG4 is a structural block diagram of a data transmission device provided in an embodiment of the present application;

FIG5 is a structural block diagram of another data transmission device provided in an embodiment of the present application;

FIG6 is a computing device provided in an embodiment of the present application;

FIG7 is a computing device cluster provided in an embodiment of the present application;

FIG8 is another computing device cluster provided in an embodiment of the present application.

DETAILED DESCRIPTION

In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the embodiments of the present application will be described in detail below in conjunction with the accompanying drawings. The terms used in the implementation method section of the present application are only used to explain the specific embodiments of the present application, and are not intended to limit the present application.

In the embodiments of the present application, "multiple" refers to two or more than two. In view of this, in the embodiments of the present application, "multiple" can also be understood as "at least two". "At least one" can be understood as one or more, for example, one, two or more. For example, including at least one means including one, two or more, and there is no restriction on which ones are included. For example, including at least one of A, B and C, then A, B, C, A and B, A and C, B and C, or A and B and C may be included. "And/or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and/or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character "/", unless otherwise specified, generally indicates that the previously associated objects are in an "or" relationship.

Unless otherwise specified, the ordinal numbers such as "first" and "second" mentioned in the embodiments of the present application are used to distinguish multiple objects and are not used to limit the order, timing, priority or importance of the multiple objects.

At present, the load balancing between the storage server and the application server in the storage system is mainly achieved by obtaining the load balancing degree between the links between each controller and the application server and/or the load balancing degree between different controllers in the storage server, and determining the load balancing degree of the overall link of the storage system through these two load balancing degrees, that is, determining the overall load balancing degree of all links in the storage system. Further, load balancing can be performed based on the overall load balancing degree to achieve load balancing between different controllers. Among them, the load balancing degree of the link is used to indicate the difference between the amount of data transmitted through different links; the load balancing degree of the controller is used to indicate the difference in the amount of data processed by different controllers. Among them, if the difference is small, it indicates that the load balancing degree is high; if the difference is large, it indicates that the load balancing degree is low. Optionally, a small difference can refer to a small difference between the highest load and the lowest load, and a large difference refers to a large difference between the highest load and the lowest load. In this way, the amount of data to be processed by each controller after load balancing is basically the same. However, if the data processing capability of a controller is reduced due to the aging of the hard disk or other phenomena, it will cause the controller to spend longer time than other controllers to process the same amount of data, and it will not be able to process the new data received in time, so the new data will be discarded, which makes the probability of retransmission of data transmitted by the link between the controller and the application server higher, affecting the performance of the storage system. For example, the storage server includes two controllers, controller 1 and controller 2, wherein controller 1 is connected to the application server through one link, and controller 2 is connected to the application server through two links. If the data processing capability of controller 1 is reduced, after the storage system is load balanced according to the overall load balancing degree of the storage system, the amount of data to be transmitted through the three links is basically the same. However, since the data processing capability of controller 1 is lower than that of controller 2, the probability of retransmission of the link between controller 1 and the application server is higher, affecting the performance of the storage system.

In view of this, in the embodiment of the present application, the storage server can detect the data processing efficiency of each of the at least two controllers included. If the data processing efficiency of a controller is lower than that of any of the other controllers, it indicates that the data processing capacity of the controller has decreased. At this time, the storage server can instruct the application server to migrate the data to be transmitted through a link corresponding to the controller to the link of other controllers for transmission, thereby reducing the amount of data to be processed by the controller with lower data processing efficiency and improving the performance of the storage system.

Please refer to Figure 1, which is a schematic diagram of a storage system provided in an embodiment of the present application. The storage system includes a storage server 10 and one or more application servers 20. The storage server 10 can provide data storage services for one or more application servers 20, and each application server 20 is installed with a client (not shown in the figure). The application server 20 can access the storage server 10 through the client to write data to the storage server 10 or read data from the storage server 10.

Optionally, different application servers 20 can access the storage server 10 through different access protocols to write and/or read data. The storage server 10 can support multiple access protocols, such as the server message block (SMB) protocol or the network file system (NFS) protocol. For example, an application server 20 with a Windows operating system can access the storage server 10 through the SMB protocol to write and/or read data; an application server 20 with an operating system of Linux or Unix can access the storage server 10 through the NFS protocol to write and/or read data.

Please refer to FIG. 2A, which is another storage system provided in an embodiment of the present application. The storage system is, for example, a detailed structure of the storage system shown in FIG. 1. The storage system includes a storage server 10 and an application server 20. Among them, the application server 20 includes at least one first port (port), and the storage server 10 includes at least two controllers, each of which includes at least one second port. Taking FIG. 2A as an example, the application server 20 includes two first ports, namely port-a and port-b, and the storage server 10 includes two controllers, namely controller A and controller B, respectively. Controller A includes two second ports, namely port1 and port2, respectively, and controller B includes two second ports, namely port3 and port4. Optionally, a first port and a second port are connected to form a link, and there may be multiple links between the application server 20 and the storage server 10, and the multiple links are, for example, obtained by connecting at least one first port and at least two second ports. Among them, the at least two second ports belong to at least two controllers. It can be understood that the first port and the second port can be physical ports or logical ports, and the embodiment of the present application does not limit the port type.

Optionally, the multiple links may be obtained by connecting a first port of the application server to at least two second ports of the controller A and the controller B. For example, there are two links between the application server 20 and the storage server 10. The two links are obtained by connecting port-a to port1 and port3 respectively. Alternatively, the multiple links may also be obtained by connecting two first ports of the application server 20 to at least two second ports of the controller A and the controller B. For example, there are two links between the application server 20 and the storage server 10. The two links are obtained by connecting port-a to port1 and port-b to port-4.

Optionally, the application server 20 may access the controller A and/or the controller B through the multiple links for data transmission. For example, the application server 20 may send a data read and write request to the controller A and/or the controller B through the multiple links to request to read data or write data. When receiving the request, the controller A or the controller B may write or read data based on the request, and send feedback information to the application server 20. For example, if the application server 20 sends a data read request to the controller A through the multiple links, the controller A reads data based on the data read request, such as reading data from a distributed file system, and sends the read data to the application server 20. If the application server 20 sends a data write request to the controller A through the multiple links, the controller A may write the data sent by the application server 20 into the distributed file system, and send feedback information to the application server 20 that the data has been written.

Optionally, when the storage server 10 provides data storage services for multiple application servers 20, the multiple application servers 20 may also be connected to the storage server 10 through a switch. For example, please refer to FIG. 2B , the storage system also includes a switch 30, each first port of the application server 20 may establish a connection with the switch 30, and the switch 30 may establish a connection with each second port of each controller of the storage server 10. At this time, each link between the application server 20 and the storage server 10 includes a first sub-link obtained by connecting the first port of the application server 20 with the switch 30 and a second sub-link obtained by connecting the switch 30 with the second port of the controller. The application server 20 may send a data read and write request to the switch 30 through the first sub-link, and the request may carry an identifier of the second port. The switch 30 may send the request to the controller through the corresponding second sub-link based on the identifier carried in the request.

In the above technical solution, the application server 20 is connected to the controller of the storage server through multiple links, so that when a link fails, for example, the port of the application server 20 or the port of the controller has hardware aging, failure, power failure or human error, the business of the failed port can be quickly switched to other ports, and when the failed port resumes working, it can take over the business switched to other ports again. In this way, the storage resources of the storage server can be fully utilized, and the stable transmission of business data can be ensured.

Based on the above content, the method provided in the embodiment of the present application is introduced below in conjunction with the drawings in the specification.

Please refer to FIG. 3 , which is a flow chart of a data transmission method provided in an embodiment of the present application. The method can be implemented by the storage system shown in FIG. 1 , FIG. 2A or FIG. 2B .

S301: When the data processing efficiency of a first controller among at least two controllers is lower than that of any controller other than the first controller among the at least two controllers, the storage server 10 sends first information to the application server 20. Accordingly, the application server 20 receives the first information from the storage server 10. The at least two controllers are part or all of the controllers included in the storage server 10.

During the use of the controller, the data processing capability of a controller may be reduced due to hardware aging and other issues. For example, the data processing efficiency of the controller may be lower than that of other controllers in the storage server 10. When the controller receives new data, it will discard the new data due to insufficient data processing capability and request the application server to resend the new data. This will cause a large number of data retransmissions on the link between the controller and the application server, which may cause link congestion and affect the performance of the storage system.

Therefore, optionally, the storage server 10 may detect each of the at least two controllers included in the storage server 10, such as timing, periodic or non-periodic detection, to determine the data processing efficiency of the at least two controllers. Thus, the storage server 10 can determine whether there is a controller whose data processing efficiency is lower than the processing efficiency of other controllers except the controller according to the detection result. Alternatively, the storage server 10 may not need to detect the controller, for example, each of the part or all of the controllers included in the storage server may send the data processing efficiency of each controller to the storage server, such as timing, periodic or non-periodic transmission, and the storage server 10 may obtain the data processing efficiency of this part or all of the controllers, thereby determining whether there is a controller whose data processing efficiency is lower than the processing efficiency of other controllers except the controller. Among them, if the data processing efficiency of the first controller is lower than any controller other than the first controller in the at least two controllers, the storage server 10 may send the first information to reduce the amount of data to be processed by the controller with lower data processing efficiency, and reduce the probability of the application server 20 retransmitting data read and write requests. Alternatively, if there is no controller whose data processing efficiency is lower than the data processing efficiency of other controllers, the storage server 10 may not send the first information.

Among them, the data processing efficiency of the first controller is lower than that of any controller other than the first controller among the at least two controllers, which can be understood as the data processing efficiency of the first controller is lower than the data processing efficiency of all controllers other than the first controller among the at least two controllers. For example, there are 3 controllers, namely controller 1 to controller 3. The data processing efficiency of controller 1 is lower than that of any controller other than controller 1, which can be understood as the data processing efficiency of controller 1 is lower than that of controller 2 and lower than that of controller 3.

Optionally, the storage server 10 may determine whether the data processing efficiency of any controller is lower than the processing efficiency of other controllers except the controller within the first time period. If the data processing efficiency of the first controller is lower than any controller of the at least two controllers except the first controller within the first time period, the storage server 10 may send the first information, otherwise the storage server 10 may not send the first information. Through the first time period, the stability of the judgment result can be increased, the probability of business switching back and forth can be reduced, and the information interaction between the storage server 10 and the application server 20 can be reduced, thereby reducing the transmission overhead.

Among them, the data processing efficiency of the first controller is lower than that of any one of the at least two controllers except the first controller within the first time period. It can be understood that the data processing efficiency of the first controller can be determined once or multiple times within the first time period, and the data processing efficiency determined each time is lower than that of any one of the at least two controllers except the first controller; or, the average value of the data processing efficiency determined this time or multiple times is lower than that of any one of the at least two controllers except the first controller.

Optionally, the data processing efficiency of the first controller is lower than that of any other controller, including one or more of the following: the difference between the amount of data processed by any other controller and the amount of data processed by the first controller within a cycle is greater than a first threshold, and the amount of data processed by any other controller is greater than the amount of data processed by the first controller; the difference between the IOPS of any other controller and the IOPS of the first controller is greater than a second threshold, and the IOPS of any other controller is greater than the IOPS of the first controller; a CPU core isolation event occurs in the first controller; the difference between a first duration used by the first controller to process a first amount of data and a second duration used by any other controller to process the first amount of data is greater than a third threshold, and the first duration is greater than the second duration; the difference between the CPU occupancy rate of the first controller and the CPU occupancy rate of any other controller is greater than a fourth threshold, and the CPU occupancy rate of the first controller is greater than the CPU occupancy rate of any other controller; the difference in IOPS of the first controller within two adjacent seconds is greater than a fifth threshold; and/or, within a preset duration, the number of times the first controller sends the second information to the application server 20 is greater than a sixth threshold, and the second information is used to request the application server 20 to resend the data read and write request.

Exemplarily, the storage server 10 may periodically count the amount of data processed by each controller in one cycle, determine the data processing efficiency of each of the at least two controllers based on the data volume, and determine whether the first controller exists based on the data processing efficiency of each controller. Optionally, when determining the data processing efficiency of each of the at least two controllers based on the data volume, the data processing efficiency of the controller determined each time in the aforementioned one or more times can be reflected by the amount of data processed in one cycle. And, the average value of the data processing efficiency determined one or more times can be reflected by the total amount of data processed in the first time period.

For example, the storage server 10 can count the amount of input/output (IO) data of each controller in one cycle, and calculate the difference in the amount of data processed by any two controllers in the cycle. If the storage server 10 calculates that the absolute value of the difference in the amount of data processed between the first controller and any other controller in the cycle is greater than a first threshold, it indicates that the data processing efficiency of the first controller is lower than that of the other controllers in the at least two controllers, that is, the first controller exists. Optionally, the amount of data processed by a controller in one cycle can also be referred to as the IO bandwidth of the controller.

The storage server 10 may also periodically count the IOPS of each controller, determine the data processing efficiency of each of the at least two controllers based on the IOPS of each controller, and determine whether the first controller exists based on the data processing efficiency of each controller.

For example, the storage server 10 may periodically obtain the number of times each controller performs data writing and reading within 1 second, calculate the total number of times each controller performs data writing and reading tasks within 1 second, and calculate the absolute value of the difference between the total number of times any two controllers perform data writing and reading tasks within 1 second. If the storage server 10 calculates that the absolute value of the difference between the total number of times the first controller and any other controller perform data writing and reading tasks within 1 second is greater than the second threshold, it indicates that the data processing efficiency of the first controller is lower than that of the other controllers of the at least two controllers, that is, the first controller exists.

The storage server 10 may also detect the IOPS of each controller within two consecutive seconds, determine the data processing efficiency of each of the at least two controllers based on the IOPS, and determine whether the first controller exists based on the data processing efficiency of each controller.

For example, the storage server 10 can calculate the difference in IOPS of each controller in two adjacent seconds. If the storage server 10 calculates that the absolute value of the difference in IOPS of the first controller in two adjacent seconds is greater than the fifth threshold, it indicates that the data processing efficiency of the first controller is lower than that of other controllers among the at least two controllers, that is, the first controller exists.

The storage server 10 may also detect whether a CPU core isolation event occurs in each controller, determine the data processing efficiency of each of the at least two controllers based on the detection result, and determine whether the first controller exists based on the data processing efficiency of each controller.

For example, a controller may include multiple CPUs, each CPU includes multiple processing cores (referred to as cores in the embodiments of the present application). When a core in a CPU fails, in order to allow the CPU to continue to work, the failed core will be isolated and not work, so that other cores can continue to work, that is, the CPU can continue to provide services. Therefore, if a core isolation event occurs in a CPU of a controller, it indicates that the CPU of the controller has fewer cores available, and the data processing capacity of the CPU of the controller is reduced, that is, there is a first controller, where the first controller is the controller corresponding to the CPU where the core isolation event occurs.

The storage server 10 can also periodically count the time taken by each controller to process the same amount of data (for example, the first amount of data), determine the data processing efficiency of each controller of the at least two controllers based on the time taken by each controller to process the first amount of data, and determine whether the first controller exists based on the data processing efficiency of each controller.

For example, the storage server 10 can obtain the time taken by each controller to process the data of the first data volume, and calculate the difference in time taken by any two controllers to process the data of the first data volume. If the storage server 10 calculates that the absolute value of the difference in time taken by the first controller and any one of the controllers to process the data of the first data volume is greater than the third threshold value, it indicates that the data processing efficiency of the first controller is lower than that of the other controllers among the at least two controllers, that is, the first controller exists. Taking the first data volume as 5G bytes and the third threshold as 200μs as an example, if the storage server 10 obtains that the time taken by the first controller to process 5GB of data is 1ms, and the time taken by any one of the controllers to process 5GB of data is 500μs, the difference in time taken by the first controller and any one of the controllers to process 5GB of data is 300μs, which is greater than 200μs, and the storage server 10 can determine that the first controller exists.

Optionally, when the time taken by the first controller to process the first data volume is longer than the time taken by any controller to process the first data volume, it indicates that the waiting time required for the data to be processed by the first controller is longer than the waiting time required for the data to be processed by any controller, and the waiting time can be referred to as IO latency.

The storage server 10 can obtain the time taken by each controller to process the first amount of data twice in succession, determine the data processing efficiency of each of the at least two controllers based on the time, and determine whether the first controller exists based on the data processing efficiency of each controller.

For example, the storage server 10 can calculate the difference in the time taken by each controller to process the first data volume twice in a row. If the storage server 10 calculates that the time taken by the first controller to process the first data volume twice in a row is greater than a threshold, it indicates that the first controller processes data sometimes fast and sometimes slow, that is, the first controller is unstable. The storage server 10 can determine the existence of the first controller.

The storage server 10 may also periodically count the CPU occupancy of each controller, determine the data processing efficiency of each of the at least two controllers based on the CPU occupancy of each controller, and determine whether the first controller exists based on the data processing efficiency of each controller.

For example, the storage server 10 can calculate the difference in CPU occupancy between any two controllers. If the absolute value of the difference in CPU occupancy between the first controller and any one of the controllers calculated by the storage server 10 is greater than the fourth threshold, it indicates that the data processing efficiency of the first controller is lower than that of the other controllers of the at least two controllers, that is, the first controller exists. Taking the fourth threshold of 5% as an example, if the CPU occupancy of the first controller obtained by the storage server 10 is 70%, the CPU occupancy of any one of the controllers is 50%, and the difference in CPU occupancy between the first controller and any one of the controllers is 20%, which is greater than 5%, the storage server 10 can determine that the first controller exists.

The storage server 10 can also periodically count the number of times each controller sends the second information to the application server 20 to instruct the application server 20 to resend data within a preset time period, determine the data processing efficiency of each controller of the at least two controllers based on the number, and determine whether the first controller exists based on the data processing efficiency of each controller.

When the data processing capability of the controller decreases, it indicates that the data processing pressure of the controller is relatively large, and the controller may discard the data received from the application server 20 and send the second information to the application server 20. Therefore, if the storage server 10 determines that the number of times the first controller sends the second information to the application server 20 within the preset time period exceeds the sixth threshold, it indicates that the data processing efficiency of the first controller is lower than that of the other controllers of the at least two controllers, that is, the first controller exists.

That is to say, the storage server 10 can determine whether it is necessary to balance the load between controllers based on the efficiency of the controllers in processing data, which helps to reduce the probability of the application server 20 retransmitting data read and write requests.

In some embodiments, the storage server 10 may also detect the load of the second port included in each controller, and/or the load of the link between each controller and the application server 20, and determine whether it is necessary to execute S301 based on the load. For example, if it is determined based on the load that the loads of different ports or links are uneven, that is, load balancing is required, the storage server 10 may execute S301, otherwise, S301 is not executed. The load includes, for example, IO bandwidth, IO latency and/or IOPS, etc. Optionally, if the IO of a second port or a link is 0 within a certain period of time, it indicates that the second port or link is faulty, such as a port network disconnection.

When the storage server 10 determines that there is a first controller, it can send the first information to the application server 20. Optionally, the storage server 10 can provide data storage services for multiple application servers 20. Therefore, each of the at least two controllers can process data from multiple application servers 20, and when the storage server 10 sends the first information to the application server 20, it can send the first information to multiple application servers 20. For example, the storage server 10 and the application server 20 are connected in a manner as shown in FIG2B , and the first link determined by the storage server 10 is a sub-link (for example, sub-link 1) among multiple second sub-links. When the storage server 10 sends the first information, it can send the first information to all application servers 20 that transmit data with the first controller through the sub-link 1, so that all application servers 20 that transmit data with the first controller through the sub-link 1 perform link switching.

Optionally, the first information may include an identifier of a link (e.g., a first link) that needs to be switched. The first link may be, for example, one link or multiple links. The first link may be randomly selected by the storage server 10 from the links corresponding to the first controller, or may be selected by the storage server 10 from the links corresponding to the first controller according to the link load. This application does not limit the number of links of the first link and the way in which the storage server 10 determines the first link. In the following embodiments, the first link including one link is taken as an example for description.

Optionally, when each controller processes data from multiple application servers 20, if the number of CPUs processing data in the controller is less than the number of application servers 20, one CPU is required to process data from multiple application servers 20. For example, there are two CPUs in controller A to process data. If the controller A needs to process data from 10 application servers 20, each CPU in the controller A can process data from five application servers 20. At this time, each CPU processes the data of the multiple application servers 20 based on the principle of time slicing. For example, after CPU1 processes the data of the first application server 20 for a certain period of time, it starts to process the data of the second application server 20, and after CPU1 processes the data of the second application server 20 for the certain period of time, it starts to process the data of the third application server 20, and after processing the data of the last application server 20, it starts to process the data of the first application server 20.

S302: The application server 20 switches the data to be transmitted through the first link to the second link.

Optionally, the second link is a link between the second controller of the at least two controllers and the application server 20. The second link may be determined by the application server 20 itself, or the second link may be specified by the storage server 10. The second link may be one link or multiple links. The embodiment of the present application does not limit the number of second links. In the following embodiment, the second link includes one link as an example for description.

If the second link is specified by the storage server 10, the storage server 10 may determine the second link in the following manner:

Mode 1: The storage server 10 may obtain the load of at least one link corresponding to a controller other than the first controller among the multiple links, select a link with the smallest link load from the at least one link, and determine the link as the second link.

Mode 2: The storage server 10 may determine the load of other controllers among the at least two controllers except the first controller, and select the controller with the smallest load from the other controllers as the target controller, and the storage server 10 may obtain the load of the link corresponding to the target controller, and select the link with the smallest link load from the links corresponding to the target controller, and determine the link as the second link. The data processing efficiency of other controllers among the at least two controllers except the first controller is substantially the same.

If the second link is determined by the application server 20 itself, the manner in which the application server 20 determines the second link may refer to the above-mentioned manner 1, which will not be described in detail herein.

Optionally, the storage server 10 may also set the state of the first link to a low priority state (also referred to as a degraded state), and send third information to the application server 20 to instruct the application server 20 to update the state of the first link to the degraded state. Therefore, when the application server 20 receives the third information, it may also set the state of the first link to the degraded state.

Optionally, after the application server 20 switches the data to be transmitted through the first link to the second link, there is no data transmission on the first link, so the storage server 10 can also perform link balancing on the link corresponding to the first controller, for example, the storage server 10 can migrate the data transmitted by other links other than the first link in the link between the first controller and the application server to the first link for transmission. Therefore, optionally, the storage server 10 can also send fourth information to at least one application server 20 that transmits data with the first controller through the other links, for instructing the at least one application server 20 to switch part of the data to be transmitted through the other links to the first link.

In the above technical solution, after the application server 20 switches the data to be transmitted through the first link to the second link, the efficiency of processing data by the at least two controllers can be made roughly the same. Take the storage server 10 as an example, which includes two controllers, namely controller 1 and controller 2, the first link is the link of controller 1, and the second link is the link of controller 2. Before balancing, the IO bandwidth of controller 1 is 5GB/S, the IO delay is 1ms, and the CPU utilization is 100%; the IO bandwidth of controller 2 is 5GB/S, the IO delay is 500us, and the CPU utilization is 50%. After balancing, the IO bandwidth of controller 1 is 3GB/S, the IO delay is 500us, and the CPU utilization is 70%; the IO bandwidth of controller 2 is 7GB/S, the IO delay is 500us, and the CPU utilization is 70%. It can be seen that after balancing, the IO delay of controller 1 and controller 2 is the same, the efficiency of controller 1 and controller 2 in processing data is the same, and the CPU utilization of controller 1 and controller 2 is also the same, which reduces the probability of application server 10 retransmitting data read and write requests and improves the performance of the storage system.

In combination with the above method embodiments, the embodiments of the present application also provide a data transmission device, which is used to execute the method executed by the storage server 10 in the above method embodiments, or to execute the method executed by the application server 20 in the above method embodiments.

Please refer to Figure 4, which is a data transmission device provided in an embodiment of the present application. The data transmission device is used to execute the method executed by the storage server 10 in the above method embodiment. The data transmission device includes a processing module 401 and a communication module 402. When the data processing efficiency of the first controller among the at least two controllers is lower than that of any controller among the at least two controllers except the first controller, the processing module 401 controls the communication module 402 to send a first information to the application server; wherein the first information is used to instruct the application server to switch the link, and the first information includes an identifier of the first link, and the first link is a link between the first controller and the application server.

In a possible implementation manner, the data processing efficiency of the first controller is lower than that of any of the controllers, including one or more of the following: the difference between the amount of data processed by any of the controllers and the amount of data processed by the first controller within a cycle is greater than a first threshold, and the amount of data processed by any of the controllers is greater than the amount of data processed by the first controller; the difference between the IOPS of any of the controllers and the IOPS of the first controller is greater than a second threshold, and the IOPS of any of the controllers is greater than the IOPS of the first controller; a CPU core isolation event occurs in the first controller; the difference between a first duration used by the first controller to process a first amount of data and a second duration used by any of the controllers to process the first amount of data is greater than a third threshold, and the first duration is greater than the second duration; the difference between the CPU occupancy rate of the first controller and the CPU occupancy rate of any of the controllers is greater than a fourth threshold, and the CPU occupancy rate of the first controller is greater than the CPU occupancy rate of any of the controllers; the difference between the IOPS of the first controller within two adjacent seconds is greater than a fifth threshold; or, within a preset duration, the number of times the first controller sends the second information to the application server is greater than a sixth threshold, and the second information is used to request the application server to resend the data read and write request.

In a possible implementation manner, the processing module 401 is further configured to: set the state of the first link to a low priority state.

In a possible implementation, the communication module 402 is further used to: send second information to the application server, where the second information is used to request the application server to update the state of the first link to the low priority state.

In a possible implementation manner, the first information is further used to indicate a second link, where the second link is a link between a second controller of the at least two controllers and the application server.

In a possible implementation, the processing module 401 is further used to: obtain load information of at least one link between the application server and a controller other than the first controller among the at least two controllers, and determine that the link with the smallest load among the at least one link is the second link; or, obtain load information of other controllers other than the first controller among the at least two controllers, determine that the controller with the smallest load is the second controller, and determine that the link with the smallest load among the links between the second controller and the application service is the second link.

Please refer to Figure 5, which is a data transmission device provided in an embodiment of the present application. The data transmission device is used to execute the method executed by the application server 20 in the above method embodiment. The data transmission device includes a communication module 501 and a processing module 502. The communication module 501 is used to receive first information from the storage server; wherein the first information is sent by the storage server when the data processing efficiency of the first controller among the at least two controllers is lower than that of any controller among the at least two controllers except the first controller, and the first information is used to instruct the application server to switch the link. The first information includes an identifier of the first link, and the first link is the link between the first controller and the application server; the processing module 502 is used to switch the data to be transmitted through the first link to the second link, and the second link is the link between the second controller among the at least two controllers and the application server.

In a possible implementation manner, the data processing efficiency of the first controller is lower than that of any of the controllers, including one or more of the following: the difference between the amount of data processed by any of the controllers and the amount of data processed by the first controller within a cycle is greater than a first threshold, and the amount of data processed by any of the controllers is greater than the amount of data processed by the first controller; the difference between the IOPS of any of the controllers and the IOPS of the first controller is greater than a second threshold, and the IOPS of any of the controllers is greater than the IOPS of the first controller; a CPU core isolation event occurs in the first controller; the difference between a first duration used by the first controller to process a first amount of data and a second duration used by any of the controllers to process the first amount of data is greater than a third threshold, and the first duration is greater than the second duration; the difference between the CPU occupancy rate of the first controller and the CPU occupancy rate of any of the controllers is greater than a fourth threshold, and the CPU occupancy rate of the first controller is greater than the CPU occupancy rate of any of the controllers; the difference between the IOPS of the first controller within two adjacent seconds is greater than a fifth threshold; or, within a preset duration, the number of times the first controller sends the second information to the application server is greater than a sixth threshold, and the second information is used to request the application server to resend the data read and write request.

In a possible implementation, the communication module 501 is further used to: receive second information from the storage server, the second information being used to request that the state of the first link be updated to a low priority state; the processing module 502 is further used to: set the state of the first link to the low priority state.

In a possible implementation manner, the first information is further used to indicate an identifier of the second link.

In a possible implementation, the processing module 502 is further used to obtain load information of at least one link between the application server and a controller other than the first controller among the at least two controllers, and determine that the link with the smallest load among the at least one link is the second link.

The present application also provides a computing device 600. As shown in FIG6 , the computing device 600 includes: a bus 601, a processor 602, a memory 603, and a communication interface 604. The processor 602, the memory 603, and the communication interface 604 communicate with each other through the bus 601. The computing device 600 can be a server or a terminal device. It should be understood that the present application does not limit the number of processors and memories in the computing device 600.

The bus 601 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, FIG. 6 is represented by only one line, but does not mean that there is only one bus or one type of bus. The bus 601 may include a path for transmitting information between various components of the computing device 600 (e.g., the memory 603, the processor 602, and the communication interface 604).

The processor 602 may include any one or more of a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor (MP), or a digital signal processor (DSP).

The memory 603 may include a volatile memory, such as a random access memory (RAM). The processor 602 may also include a non-volatile memory, such as a read-only memory (ROM), a flash memory, a hard disk drive (HDD), or a solid state drive (SSD).

The memory 603 stores executable program codes, and the processor 602 executes the executable program codes to respectively implement the functions of the aforementioned processing module 401 and the communication module 402, or to implement the functions of the aforementioned communication module 501 and the processing module 502, thereby implementing the data transmission method. That is, the memory 603 stores instructions for executing the data transmission method.

Alternatively, the memory 603 stores executable codes, and the processor 602 executes the executable codes to respectively implement the functions of the aforementioned storage server and application server, thereby implementing the data transmission method. That is, the memory 603 stores instructions for executing the data transmission method.

The communication interface 604 uses a transceiver module such as, but not limited to, a network interface card or a transceiver to implement communication between the computing device 600 and other devices or a communication network.

The embodiment of the present application also provides a computing device cluster. The computing device cluster includes at least one computing device. The computing device can be a server, such as a central server, an edge server, or a local server in a local data center. In some embodiments, the computing device can also be a terminal device such as a desktop computer, a laptop computer, or a smart phone.

As shown in Fig. 7, the computing device cluster includes at least one computing device 600. The memory 603 in one or more computing devices 600 in the computing device cluster may store the same instructions for executing the data transmission method.

In some possible implementations, the memory 603 of one or more computing devices 600 in the computing device cluster may also store partial instructions for executing the data transmission method. In other words, the combination of one or more computing devices 600 may jointly execute instructions for executing the data transmission method.

It is understandable that the memory 603 in different computing devices 600 in the computing device cluster can store different instructions, which are respectively used to execute part of the functions of the storage server. That is, the instructions stored in the memory 603 in different computing devices 600 can implement the functions of one or more modules in the application server and the storage server.

In some possible implementations, one or more computing devices in the computing device cluster can be connected via a network. Wherein, the network can be a wide area network or a local area network, etc. FIG. 8 shows a possible implementation. As shown in FIG. 8 , two computing devices 600A and 600B are connected via a network. Specifically, the communication interface in each computing device is connected to the network. In this type of possible implementation, the memory 603 in the computing device 600A stores instructions for executing the function of the storage server.

It should be understood that the functions of the computing device 600A shown in FIG8 may also be completed by multiple computing devices 600. Similarly, the functions of the computing device 600B may also be completed by multiple computing devices 600.

The embodiment of the present application also provides another computing device cluster. The connection relationship between the computing devices in the computing device cluster can be similar to the connection mode of the computing device cluster described in Figures 7 and 8. The difference is that the memory 603 in one or more computing devices 600 in the computing device cluster can store the same instructions for executing the data transmission method.

In some possible implementations, the memory 603 of one or more computing devices 600 in the computing device cluster may also store partial instructions for executing the data transmission method. In other words, the combination of one or more computing devices 600 may jointly execute instructions for executing the data transmission method.

It is understandable that the memory 603 in different computing devices 600 in the computing device cluster can store different instructions for executing part of the functions of the storage system. That is, the instructions stored in the memory 603 in different computing devices 600 can implement the functions of one or more devices in the application server and the storage server.

The embodiment of the present application also provides a computer program product including instructions. The computer program product may be software or a program product including instructions that can be run on a computing device or stored in any available medium. When the computer program product is run on at least one computing device, the at least one computing device executes the data transmission method.

The embodiment of the present application also provides a computer-readable storage medium. The computer-readable storage medium can be any available medium that can be stored by a computing device or a data storage device such as a data center containing one or more available media. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state hard disk). The computer-readable storage medium includes instructions that instruct the computing device to execute the data transmission method.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the protection scope of the technical solutions of the embodiments of the present invention.

Claims (15)

1. A data transmission method, applied to a storage system, the storage system including a storage server and an application server, the storage server including a plurality of controllers, the application server being connected to at least two controllers of the plurality of controllers, the method comprising:

When the data processing efficiency of a first controller in the at least two controllers is lower than that of any one controller except the first controller in the at least two controllers, the storage server sends first information to the application server; the first information is used for indicating the application server to switch links, and the first information comprises an identifier of a first link, wherein the first link is a link between the first controller and the application server;

the application server switches data to be transmitted through the first link to a second link between a second controller of the at least two controllers and the application server.

2. The method of claim 1, wherein the first controller has a lower data processing efficiency than the any one of the controllers, comprising one or more of:

The difference between the data amount processed by any one of the controllers and the data amount processed by the first controller in one period is larger than a first threshold value, and the data amount processed by any one of the controllers is larger than the data amount processed by the first controller;

The difference value between the read-write times per second IOPS of any one controller and the IOPS of the first controller is larger than a second threshold value, and the IOPS of any one controller is larger than the IOPS of the first controller;

the first controller generates a Central Processing Unit (CPU) core isolation event;

A difference between a first time period for the first controller to process a first amount of data and a second time period for the any one of the controllers to process the first amount of data is greater than a third threshold, and the first time period is greater than the second time period;

the difference value between the CPU occupancy rate of the first controller and the CPU occupancy rate of any one of the controllers is larger than a fourth threshold value, and the CPU occupancy rate of the first controller is larger than the occupancy rate of the CPU of any one of the controllers;

the difference value of the IOPS in two seconds adjacent to the first controller is larger than a fifth threshold value; or alternatively, the first and second heat exchangers may be,

And within a preset duration, the number of times of sending second information to the application server by the first controller is larger than a sixth threshold, and the second information is used for requesting the application server to resend a data read-write request.

3. The method of claim 1 or 2, wherein the method further comprises:

the storage server sets the state of the first link to a low priority state.

4. A method as claimed in claim 3, wherein the method further comprises:

the storage server sends third information to the application server, wherein the third information is used for indicating the application server to update the state of the first link to the low-priority state;

the application server sets the state of the first link to the low priority state.

5. The method of any one of claims 1-4, further comprising:

The application server determines a link with the minimum link load as the second link; or alternatively, the first and second heat exchangers may be,

The first information is also used to indicate the second link.

6. The method of claim 5, wherein the first information is further for indicating the second link, the method further comprising:

The storage server acquires load information of at least one link between the application server and a controller except the first controller in the at least two controllers, and determines a link with the smallest load in the at least one link as the second link; or alternatively

The storage server acquires load information of other controllers except the first controller in the at least two controllers, determines the controller with the smallest load as the second controller, and determines the link with the smallest load in the links between the second controller and the application server as the second link.

7. A storage system comprising a storage server comprising a plurality of controllers and an application server coupled to at least two of the plurality of controllers, wherein,

The storage server is used for sending first information to the application server when the data processing efficiency of a first controller in the at least two controllers is lower than that of any one controller except the first controller in the at least two controllers; the first information is used for indicating the application server to switch links, and the first information comprises an identifier of a first link, wherein the first link is a link between the first controller and the application server;

The application server is configured to switch data to be transmitted through the first link to a second link, where the second link is a link between a second controller of the at least two controllers and the application server.

8. The storage system of claim 7, wherein the first controller has a lower data processing efficiency than the any one of the controllers, comprising one or more of:

The difference between the data amount processed by any one of the controllers and the data amount processed by the first controller in one period is larger than a first threshold value, and the data amount processed by any one of the controllers is larger than the data amount processed by the first controller;

The difference value between the read-write times per second IOPS of any one controller and the IOPS of the first controller is larger than a second threshold value, and the IOPS of any one controller is larger than the IOPS of the first controller;

the first controller generates a Central Processing Unit (CPU) core isolation event;

A difference between a first time period for the first controller to process a first amount of data and a second time period for the any one of the controllers to process the first amount of data is greater than a third threshold, and the first time period is greater than the second time period;

the difference value between the CPU occupancy rate of the first controller and the CPU occupancy rate of any one of the controllers is larger than a fourth threshold value, and the CPU occupancy rate of the first controller is larger than the occupancy rate of the CPU of any one of the controllers;

the difference value of the IOPS in two seconds adjacent to the first controller is larger than a fifth threshold value; or alternatively, the first and second heat exchangers may be,

And within a preset duration, the number of times of sending second information to the application server by the first controller is larger than a sixth threshold, and the second information is used for requesting the application server to resend a data read-write request.

9. The storage system of claim 7 or 8, wherein the storage server is further configured to:

The state of the first link is set to a low priority state.

10. The storage system of claim 9, wherein the memory device comprises a memory device,

The storage server is further configured to: transmitting third information to the application server, wherein the third information is used for indicating the application server to update the state of the first link to the low-priority state;

the application server is further configured to: the state of the first link is set to the low priority state.

11. The storage system of any one of claim 7 to 10,

The application server is configured to determine that a link with a minimum link load is the second link; or alternatively, the first and second heat exchangers may be,

The first information is also used to indicate the second link.

12. The storage system of claim 11, wherein the first information is further for indicating the second link, the storage server is further for:

Load information of at least one link between the application server and a controller except the first controller in the at least two controllers is obtained, and a link with the minimum load in the at least one link is determined to be the second link; or alternatively

And acquiring load information of other controllers except the first controller in the at least two controllers, determining the controller with the smallest load as the second controller, and determining the link with the smallest load in the links between the second controller and the application service as the second link.

13. A cluster of computing devices, comprising at least one computing device, each computing device comprising a processor and a memory;

A processor of the at least one computing device is configured to execute instructions stored in a memory of the at least one computing device to cause the cluster of computing devices to perform the method performed by the storage server or the application server as claimed in any one of claims 1 to 6.

14. A computer program product containing instructions that, when executed by a cluster of computing devices, cause the cluster of computing devices to perform the method performed by the storage server or the application server of any one of claims 1-6.

15. A computer readable storage medium comprising computer program instructions which, when executed by a cluster of computing devices, perform a method performed by a storage server or the application server as claimed in any one of claims 1 to 6.