Traffic scheduling method and device for multi-data center system, storage medium and program product

Abstract

The invention discloses a traffic scheduling method and device for a multi-data center system, a storage medium and a program product, and relates to the technical field of data processing. According to the method, when the edge node receives the traffic data of the user side, the unique token value corresponding to the traffic data is obtained, and if the unique token value does not have a specified data center, the scheduling proportion of each data center is determined based on the bandwidth occupation proportion, the bandwidth margin and the data packet round-trip time of each data center, and the scheduling of the data centers is completed. And then determining a target data center based on the scheduling proportion of the data center, and forwarding the traffic data to the target data center. According to the method, the scheduling proportion is dynamically adjusted according to the real-time bandwidth data and the data packet round-trip time data, it is ensured that resources are reasonably distributed among the data centers, and the problem that a current multi-data-center system architecture cannot schedule flow requests in real time is solved.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a traffic scheduling method, device, storage medium and program product for a multi-data center system. Background Art

[0002] Multi-data center architecture is a high-availability system architecture that achieves business continuity and disaster recovery by deploying the same application system and data in data centers with different geographical locations. Traffic data can be distributed to each data center through a load balancer to achieve load balancing and performance optimization.

[0003] The current multi-data center system architecture mainly distributes traffic requests to the primary data center and the backup data center according to a specified ratio by building an nginx proxy inside the primary data center. However, when the network condition of the primary data center is poor or the communication rate is low, the above multi-data center system architecture cannot flexibly schedule user traffic requests, and when the network entrance or isolation area of ​​the primary data center fails, the application system of the primary data center will not be able to provide services to the outside world. The operation and maintenance personnel need to modify the domain name resolution configuration of the primary data center and point the domain name to the backup data center so that the backup data center can continue to provide services. However, the above process requires a certain amount of operation time and domain name resolution effectiveness time, which will cause service interruption and cannot guarantee the continuity of traffic request business. Therefore, the current multi-data center system architecture has the problem of being unable to schedule traffic requests in real time.

[0004] The above contents are only used to assist in understanding the technical solution of the present application and do not constitute an admission that the above contents are prior art. Summary of the invention

[0005] The main purpose of this application is to provide a traffic scheduling method, device, storage medium and program product for a multi-data center system, aiming to solve the technical problem that the current multi-data center system architecture cannot schedule traffic requests in real time.

[0006] To achieve the above objectives, the present application proposes a traffic scheduling method for a multi-data center system, which is applied to an edge node. The method includes: When the edge node receives the flow data from the user side, obtaining a unique token value corresponding to the flow data; If the unique token value does not specify the data center, determine the scheduling ratio of each data center based on the bandwidth occupancy ratio, bandwidth margin and data packet round trip time of each data center; Based on the scheduling ratio of the data center, a target data center is determined, and the traffic data is forwarded to the target data center.

[0007] In one embodiment, the step of determining the scheduling ratio of each data center based on the bandwidth occupancy ratio, bandwidth margin and data packet round trip time of each data center includes: Obtaining the bandwidth occupancy ratio and bandwidth margin of each of the data centers within a preset time interval; Determine the round trip time ratio difference of the data packets between the data centers based on the round trip time of the data packets of the data centers; Obtaining a scheduling benchmark value for each of the data centers; The scheduling ratio of each of the data centers is determined based on the bandwidth occupancy ratio, the bandwidth margin, the scheduling reference value, and the data packet round-trip time ratio difference of each of the data centers.

[0008] In one embodiment, the step of determining the scheduling ratio of each data center based on the bandwidth occupancy ratio, the bandwidth margin, the scheduling reference value, and the data packet round-trip time ratio difference of each data center includes: If the bandwidth occupancy ratio of each of the data centers is less than or equal to the preset occupancy ratio, and the data packet round-trip time ratio difference is less than or equal to the preset round-trip time ratio difference, determining the scheduling ratio of each of the data centers according to the scheduling reference value; If the bandwidth occupancy ratio of each of the data centers is less than or equal to the preset occupancy ratio, and the data packet round-trip time ratio difference is greater than the preset round-trip time ratio difference, the scheduling ratio of each of the data centers is determined according to the scheduling reference value and the data packet round-trip time ratio difference.

[0009] In one embodiment, the step of determining the scheduling ratio of each data center based on the bandwidth occupancy ratio, the bandwidth margin, the scheduling reference value, and the data packet round trip time ratio difference of each data center further includes: If the bandwidth occupancy ratio of each of the data centers is greater than the preset occupancy ratio, and the round-trip time ratio difference of the data packets is less than or equal to the preset round-trip time ratio difference, determining the scheduling ratio of each of the data centers according to the scheduling reference value and the bandwidth margin; If the bandwidth occupancy ratio of each of the data centers is greater than the preset occupancy ratio, and the data packet round-trip time ratio difference is greater than the preset round-trip time ratio difference, the scheduling ratio of each of the data centers is determined according to the scheduling reference value, the bandwidth margin and the data packet round-trip time ratio difference.

[0010] In one embodiment, the step of determining the scheduling ratio of each data center based on the bandwidth occupancy ratio, the bandwidth margin, the scheduling reference value, and the data packet round trip time ratio difference of each data center further includes: If there is a data center with a bandwidth occupancy ratio greater than the preset occupancy ratio, the data center with a bandwidth occupancy ratio greater than the preset occupancy ratio is used as the first data center, and the data center with a bandwidth occupancy ratio less than or equal to the preset occupancy ratio is used as the second data center; If the round-trip time ratio difference of the data packets between the first data center and the second data center is less than or equal to the preset round-trip time ratio difference, determining that the call ratio of the second data center is a preset multiple of the corresponding scheduling reference value; If the round-trip time ratio difference of the data packets between the first data center and the second data center is greater than the preset round-trip time ratio difference, determining the call ratio of the second data center to be a preset multiple of the corresponding scheduling reference value and the round-trip time ratio difference of the data packets; Based on the calling ratio of the second data center, the corresponding remaining traffic scheduling ratio is determined to be the calling ratio of the first data center.

[0011] In one embodiment, the step of determining the round trip time ratio difference of data packets between the data centers based on the round trip time of the data packets of the data centers comprises: Obtaining historical round-trip time of traffic data from each of the data centers; In a first preset time period, respectively calculating the average of the historical round-trip time of the traffic data in each of the data centers; In the second preset time period, exponential smoothing calculation is performed on the mean of the historical round-trip time of each data center to obtain the round-trip time of the data packet of each data center.

[0012] In one embodiment, the method further comprises: If the unique token value corresponding to the traffic data exists in the designated data center, forwarding the traffic data to the designated data center; Alternatively, the traffic data is forwarded to the historically corresponding data center according to the unique token value associated with the traffic data and the historical corresponding relationship between the data centers.

[0013] In addition, to achieve the above-mentioned objectives, the present application also proposes a traffic scheduling device for a multi-data center system, the device comprising: a memory, a processor, and a computer program stored on the memory and executable on the processor, the computer program being configured to implement the steps of the traffic scheduling method for the multi-data center system as described above.

[0014] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by the processor, the steps of the traffic scheduling method for the multi-data center system as described above are implemented.

[0015] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, it implements the steps of the traffic scheduling method for the multi-data center system as described above.

[0016] The present application proposes a traffic scheduling method for a multi-data center system. When the edge node receives the traffic data from the user side, it obtains the unique token value corresponding to the traffic data. If the unique token value does not exist in the specified data center, the scheduling ratio of each data center is determined based on the bandwidth occupancy ratio, bandwidth margin and data packet round-trip time of each data center. Then, based on the scheduling ratio of the data center, the target data center is determined and the traffic data is forwarded to the target data center. The above method dynamically adjusts the scheduling ratio according to the real-time bandwidth data and data packet round-trip time data to ensure that resources are reasonably allocated between data centers, solving the problem that the current multi-data center system architecture cannot schedule traffic requests in real time. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0019] Figure 1 A schematic diagram of a dual-service architecture of a traffic data scheduling system involved in a traffic scheduling method for a multi-data center system in an embodiment of the present application; Figure 2 A flow chart of a method for traffic scheduling in a multi-data center system according to an embodiment of the present invention; Figure 3 A flow chart of a second embodiment of a traffic scheduling method for a multi-data center system of the present application; Figure 4 A flow chart of a method for traffic scheduling in a multi-data center system according to a third embodiment of the present invention; Figure 5A flow chart of a fourth embodiment of a traffic scheduling method for a multi-data center system of the present application; Figure 6 A flow chart of a method for traffic scheduling in a multi-data center system according to a fifth embodiment of the present invention; Figure 7 A flow chart of the seventh embodiment of the method for traffic scheduling in a multi-data center system of the present application; Figure 8 Schematic diagram of the device structure of the hardware operating environment involved in the traffic scheduling method for a multi-data center system in an embodiment of the present application.

[0020] The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0021] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0022] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0023] The current dual-data center system architecture mainly distributes traffic requests to the primary data center and the backup data center according to a specified ratio by building an nginx proxy inside the primary data center. However, when the network condition of the primary data center is poor or the communication rate is low, the dual-data center system architecture cannot flexibly schedule user traffic requests. When the network entrance or isolation area of ​​the primary data center fails, the application system of the primary data center will not be able to provide services to the outside world. The operation and maintenance personnel need to modify the domain name resolution configuration of the primary data center and point the domain name to the backup data center so that the backup data center can continue to provide services. However, the above process requires a certain amount of operation time and domain name resolution effectiveness time, which will cause service interruption and cannot guarantee the continuity of traffic request business. Therefore, the current multi-data center system architecture has the problem of being unable to schedule traffic requests in real time.

[0024] In view of the above problems, the present application proposes a traffic scheduling method for a multi-data center system. When the edge node receives the traffic data from the user side, it obtains the unique token value corresponding to the traffic data. If the unique token value does not exist in the specified data center, the scheduling ratio of each data center is determined based on the bandwidth occupancy ratio, bandwidth margin and data packet round-trip time of each data center. Then, based on the scheduling ratio of the data center, the target data center is determined and the traffic data is forwarded to the target data center. The above method dynamically adjusts the scheduling ratio according to the real-time bandwidth data and data packet round-trip time data to ensure that resources are reasonably allocated between data centers, solving the problem that the current multi-data center system architecture cannot schedule traffic requests in real time.

[0025] It should be noted that the execution entity in the embodiments of the present application can be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an electronic device that can realize the above functions.

[0026] For example, in order to more clearly illustrate this technical solution, this application is described using a traffic data scheduling system with a dual data center and dual service architecture as the execution subject. Figure 1 , Figure 1 Schematic diagram of the dual service architecture of the traffic data scheduling system involved in the following embodiments of the present application.

[0027] Specifically, an edge node traffic proxy module is introduced between the user side and the data center side. The edge node traffic proxy module dispatches the business data traffic request initiated by the user to the data center, and then the nginx proxy in the internal isolation area of ​​the data center sends the business data traffic to the data center API (Application Programming Interface) gateway. Among them, a parameter control module is set in the data center, which is responsible for collecting configuration parameters from the data center and sending the configuration parameters to the edge node traffic proxy module. The parameter control module can obtain the correspondence between the application system domain name and the public network IP (Internet Protocol) address of the application system providing services to the outside from the domain name system; at the same time, it can also obtain the bandwidth data of the data center through the computer room traffic monitoring module of the data center. A unified user center module is also set in the core area of ​​the data center for registering and managing user information, and generating a unique token value according to the user's registered account. The edge node traffic proxy module will store the correspondence between the user's unique token value and the data center, so that when the user initiates a traffic request again, the edge node traffic proxy module can dispatch the user's traffic request to the same data center to maintain the continuity of business services.

[0028] Based on this, the first embodiment proposed in this application provides a traffic scheduling method for a multi-data center system, referring to Figure 2 In this embodiment, the traffic scheduling method of the multi-data center system includes steps S10 to S30: Step S10, when the edge node receives the traffic data from the user side, obtains the unique token value corresponding to the traffic data.

[0029] It should be noted that the token value is a string or number used to identify the user's identity and is associated with the user's account. When a user initiates a request, the unified user center module can generate a unique token based on the user's account identifier.

[0030] For example, suppose user A registers an account for the first time on a video-on-demand application and requests to watch a video. The unified user center module will convert the account ID (Identification) registered by user A on the video-on-demand application into an output sequence value of fixed length. For example, a one-way hash algorithm is used to map the user's 8-digit account ID "10132334" to a 32-bit string token "ACVBNMZXCVBFGTYIKFGCBDOUNSLBGEQL".

[0031] Optionally, in addition to the one-way hash algorithm, other encodings such as Base64 encoding, HMAC (Hash-based Message Authentication Code), UUID (Universally Unique Identifier), etc. can be used to convert the user account into a unique token.

[0032] Step S20: If the unique token value does not specify a data center, the scheduling ratio of each data center is determined based on the bandwidth occupancy ratio, bandwidth margin and data packet round-trip time of each data center.

[0033] It should be noted that the bandwidth occupancy ratio refers to the ratio of the bandwidth currently used by the data center to its total bandwidth capacity. For example, if the total bandwidth of a data center is 100Mbps and 80Mbps is currently used, the bandwidth occupancy ratio is 80%. The bandwidth margin refers to the amount of bandwidth remaining available in the data center. In the above example, if the total bandwidth is 100Mbps and 80Mbps has been used, then the bandwidth margin is 20Mbps. The round-trip time of the data packet can reflect the transmission speed of the traffic data inside and outside the data center. It can be determined by the time when the traffic data is allocated from the edge node traffic proxy module to the data center for processing and then the requested data is returned to the edge node traffic proxy module, reflecting the network performance of the data center. The scheduling ratio of the data center is calculated by the edge node according to the preset traffic scheduling algorithm, which is used to indicate the ratio at which the traffic data should be forwarded to each data center when the traffic data does not exist in the specified data center.

[0034] Optionally, the traffic data scheduling system continuously monitors the bandwidth occupancy ratio, bandwidth margin and data packet round trip time of each data center, and regularly evaluates the current status of each data center based on the above performance indicators, such as every minute or every hour, and comprehensively determines the traffic ratio that each data center should bear through these performance indicators. Among them, if the bandwidth occupancy ratio of a data center is very high, the traffic data scheduling system may reduce its scheduling ratio to avoid overload; if the bandwidth margin of a data center is large, the traffic data scheduling system may increase its scheduling ratio to make full use of resources; if the data packet round trip time of a data center is short, it means that the data center has better network quality or higher communication rate, and the traffic data scheduling system may tend to schedule more traffic data to the data center to improve the overall response speed.

[0035] Dynamically adjusting the scheduling ratio according to actual performance can more effectively utilize the resources of each data center and ensure that traffic data is distributed to the data center with the best performance in real time, thereby improving the response speed and quality of the service.

[0036] Step S30: determining a target data center based on the scheduling ratio of the data center, and forwarding the traffic data to the target data center.

[0037] Optionally, when a user logs in to the application for the first time and initiates traffic data for a specific network domain name, the token can be placed in the custom header of the http (Hypertext Transfer Protocol) request and sent to the edge node traffic proxy module, and the edge node traffic proxy module forwards the traffic data to the corresponding data center according to the calculated scheduling ratio of the data center. Among them, before forwarding the traffic data, the edge node traffic proxy module can obtain the corresponding relationship between the user's unique token and the public IP address of the application system providing services to the outside through the parameter control module, so that when the edge traffic proxy node obtains the user's traffic data, it can quickly establish a connection with the application system according to the corresponding public IP address.

[0038] Exemplarily, the parameter control module in the data center can query the domain name system for the public IP address corresponding to the application system domain name within a preset time interval. After obtaining the public IP address, the parameter control module sends the public IP address to the edge node traffic proxy module. The edge node traffic proxy module uses the received public IP address to directly establish a network connection with the application system to realize the application scheduling of traffic data.

[0039] To better understand the solution provided in this example, this example is further explained in combination with specific application scenarios.

[0040] Assume that an online video streaming service application needs to provide video content through a data center. For load balancing and disaster recovery, the service operator sets up two data centers X and Y. Both data centers X and Y have the ability to process video streaming requests, and their scheduling ratios are 30% and 70% respectively. This means that in the absence of specific allocation, 30% of the traffic data will be scheduled to data center X, and 70% of the traffic data will be scheduled to data center Y.

[0041] When a user enters a username and password to register an account in the user interface of the online video streaming service application, the unified user center module of the data center generates and saves a unique token based on the user's registered account, assuming it is "1234567". After the user logs in to the online video streaming service application, he wants to access a service of the online video streaming service application, which is deployed on a specific network domain name, such as "https: / / service.example.com". The application puts the generated token in the custom header of the HTTP request, such as "X-User-Token:1234567". The edge node traffic proxy module receives the traffic data and extracts the token in the custom header. The edge node traffic proxy module determines the public IP address corresponding to "https: / / service.example.com", such as "123.123.1.100", based on the correspondence between the token and the public IP address of the application system providing services to the outside world, and establishes a network connection with the online video streaming service application service through the public IP address.

[0042] In this embodiment, through tokens and dynamic scheduling ratios, the traffic data scheduling system can flexibly handle the request distribution between different users and different data centers, adapt to changing network conditions and user needs, and when a data center fails, transfer requests to other data centers to achieve real-time scheduling of traffic requests and improve the reliability and fault tolerance of the traffic data scheduling system.

[0043] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar contents as those in the above-mentioned embodiment 1 can refer to the above introduction, and will not be repeated later. Figure 3 , step S20 also includes steps S21 to S24: Step S21, obtaining the bandwidth occupancy ratio and bandwidth margin of each data center within a preset time interval.

[0044] Optionally, the edge node traffic proxy module obtains the current bandwidth occupancy ratio and bandwidth margin from the traffic monitoring module of the data center at a preset time interval, such as every minute.

[0045] Exemplarily, in the traffic data scheduling system, the parameter control module of the data center can periodically collect the current network traffic and computer room exit bandwidth of the computer room exit router from the computer room traffic monitoring module inside the data center, and calculate the bandwidth occupancy ratio and bandwidth margin of the current computer room exit router based on the current network traffic and computer room exit bandwidth, that is, bandwidth occupancy ratio = current network traffic / computer room exit bandwidth; bandwidth margin = computer room exit bandwidth - current network traffic.

[0046] Optionally, use different traffic monitoring tools such as SNMP (Simple Network Management Protocol) and NetFlow to obtain bandwidth information, or set the frequency of obtaining bandwidth information to real-time to more accurately understand the network status of the data center.

[0047] Step S22: determining the round trip time ratio difference of the data packet between the data centers based on the round trip time of the data packet of each data center.

[0048] Exemplarily, the edge node traffic proxy module periodically sends data packets of speed measurement requests from the parameter control module to the two data centers, records the round-trip time of the data packets of the two data centers, and calculates the round-trip time ratio difference of the data packets between the two data centers for decision-making of the traffic scheduling algorithm. The edge node traffic proxy module can adjust the distribution ratio of user traffic between the two data centers according to the round-trip time ratio difference of the data packets, and schedule the user traffic to the data center with higher communication efficiency to improve the user experience.

[0049] Step S23, obtaining the scheduling reference value of each of the data centers.

[0050] Step S24, determining the scheduling ratio of each of the data centers based on the bandwidth occupancy ratio, the bandwidth margin, the scheduling reference value, and the data packet round-trip time ratio difference of each of the data centers.

[0051] It should be noted that the dispatching benchmark value is a reference dispatching ratio value pre-set by business personnel according to business needs. For example, it is pre-set to allocate more traffic data to data centers with larger bandwidth capacity, or different dispatching benchmark values ​​are set according to the characteristics of different businesses, for example, more traffic data of high-priority businesses is allocated to the main data center to ensure the processing efficiency of key businesses.

[0052] It should also be noted that the scheduling ratio refers to the ratio of traffic data actually allocated to the data center after calculation by the edge node traffic proxy module, which is calculated by the edge node traffic proxy module based on the collected data center parameters using a preset traffic scheduling algorithm.

[0053] Optionally, business personnel set the scheduling benchmark value of the data center in the parameter control module according to business needs. The edge node traffic proxy module regularly obtains the scheduling benchmark value of the data center from the parameter control module, and dynamically adjusts the distribution ratio of traffic data between the two data centers based on the obtained scheduling benchmark value, real-time bandwidth information, data packet round-trip time and other factors to achieve load balancing and intelligent scheduling.

[0054] In this implementation, by monitoring bandwidth occupancy ratio, bandwidth margin, and data packet round trip time, the traffic data scheduling system can intelligently schedule traffic data to a data center with higher communication efficiency. In addition, by setting a scheduling benchmark value and dynamically adjusting the scheduling ratio, and comprehensively considering bandwidth information, data packet round trip time, and scheduling benchmark values, the traffic data scheduling system can make more accurate traffic scheduling decisions and quickly adapt to changes in business needs.

[0055] Based on the above embodiments of the present application, in the third embodiment of the present application, the same or similar contents as those in the above embodiments can be referred to the above introduction, and will not be described in detail later. Figure 4 , step S24 also includes steps S241-S242: Step S241, if the bandwidth occupancy ratio of each data center is less than or equal to the preset occupancy ratio, and the data packet round-trip time ratio difference is less than or equal to the preset round-trip time ratio difference, determine the scheduling ratio of each data center according to the scheduling reference value.

[0056] For example, it is assumed that the bandwidth occupancy ratio is , where the bandwidth usage ratio of the first data center A is recorded as , the bandwidth usage ratio of the second data center B is recorded as According to business needs, the benchmark values ​​of different data center traffic scheduling ratios are set in the parameter control module as , where the scheduling benchmark value of data center A is recorded as , the scheduling benchmark value of data center B is recorded as Assume that the round trip time of the data packet in data center A is , the round trip time of the data packet in data center B is , the round trip time ratio difference of its data packets is | - |.

[0057] When the bandwidth usage ratio of the two data centers is less than or equal to the preset usage ratio, which is set to 90%, and the round-trip time ratio difference of the data packets of the two data centers is less than or equal to the preset round-trip time ratio difference, for example, 30%, that is, 90%, 90% and | - | When , it means that both data centers have sufficient bandwidth margin, and because the difference in the round-trip time of data packets between the two data centers is relatively small, the traffic data can be distributed relatively evenly. At this time, the scheduling ratio is determined according to the scheduling benchmark value of the data center. For example, the scheduling ratio of data center A is determined to be = , the scheduling ratio of data center B is =1- .

[0058] Step S242, if the bandwidth occupancy ratio of each of the data centers is less than or equal to the preset occupancy ratio, and the data packet round-trip time ratio difference is greater than the preset round-trip time ratio difference, determine the scheduling ratio of each of the data centers based on the scheduling reference value and the data packet round-trip time ratio difference.

[0059] Exemplarily, when the bandwidth occupancy ratios of the two data centers are both less than or equal to the preset occupancy ratio, which is set to 90%, and the round-trip time ratio difference of the data packets of the two data centers is greater than the preset round-trip time ratio difference, such as 30%, for example 90%, 90% and | - | When , it means that both data centers have sufficient bandwidth margin, but the difference in the round-trip time of the data packets of the two data centers is relatively large. In this case, the scheduling ratio of the two data centers is appropriately adjusted according to the difference in the round-trip time of the data packets of the two data centers, and the scheduling ratio of the data center with a shorter round-trip time of the data packets is increased. For example, assuming that the round-trip time of the data packets of data center A is shorter, the scheduling ratio of data center A is increased according to the difference in the round-trip time ratio of the data packets of the two data centers. The scheduling ratio of data center A is determined to be = | - | / 2, then the scheduling ratio of data center B is =1- .

[0060] It should be noted that the values ​​of the preset occupancy ratio and the preset round-trip time ratio difference and the specific formula in this embodiment can be adjusted according to actual conditions, and this application does not make specific limitations on this.

[0061] Based on the above embodiments of the present application, in the fourth embodiment of the present application, the same or similar contents as those in the above embodiments can be referred to the above introduction, and will not be described in detail later. Figure 5 , step S24 also includes steps S243~S244: Step S243, if the bandwidth occupancy ratio of each of the data centers is greater than the preset occupancy ratio, and the data packet round-trip time ratio difference is less than or equal to the preset round-trip time ratio difference, determine the scheduling ratio of each of the data centers according to the scheduling reference value and the bandwidth margin.

[0062] For example, it is assumed that the bandwidth margin is , assuming that the bandwidth margin of the first data center A is , the bandwidth margin of the second data center is recorded as .

[0063] When the bandwidth usage ratios of the two data centers are both greater than the preset usage ratio, which is set to 90%, and the round-trip time ratio difference of the data packets of the two data centers is less than or equal to the preset round-trip time ratio difference, such as 30%, for example 90%, 90% and | - | When , it means that the bandwidth margin of the two data centers is relatively tight, and the round-trip time ratio of the data packets in the two data centers is not much different, the traffic data will be allocated to the data center with more bandwidth margin. For example, assuming that the bandwidth margin of data center A is If it is relatively large, the scheduling ratio of data center A is determined =( / ) , the scheduling ratio of data center B =1- .

[0064] Step S244, if the bandwidth occupancy ratio of each of the data centers is greater than the preset occupancy ratio, and the data packet round-trip time ratio difference is greater than the preset round-trip time ratio difference, determine the scheduling ratio of each of the data centers based on the scheduling reference value, the bandwidth margin and the data packet round-trip time ratio difference.

[0065] For example, when the bandwidth occupancy ratios of the two data centers are both greater than the preset occupancy ratio, which is set to 90%, and the round-trip time ratio difference of the data packets of the two data centers is greater than the preset round-trip time ratio difference, such as 30%, for example 90%, 90% and | - | When , it means that the bandwidth margin of the two data centers is relatively tight, and the round-trip time ratio of the two data centers is quite different, the scheduling ratio of the two data centers is adjusted according to the difference in the round-trip time ratio of the two data centers. For example, if the round-trip time of the data packets of data center A is shorter, the scheduling ratio of data center A is increased according to the difference in the round-trip time ratio of the data packets. =( / ) , the scheduling ratio of data center B is =1- .

[0066] For example, when determining the first scheduling ratio After that, the remaining traffic scheduling ratio is used as the second scheduling ratio ,Right now =1- .

[0067] It should be noted that the values ​​of the preset occupancy ratio and the preset round-trip time ratio difference and the specific formula in this embodiment can be adjusted according to actual conditions, and this application does not make specific limitations on this.

[0068] Based on the above embodiments of the present application, in the fifth embodiment of the present application, the same or similar contents as those in the above embodiments can be referred to the above introduction, and will not be described in detail later. Figure 6 , step S24 also includes steps S245 to S248: Step S245: If there is a data center with a bandwidth occupancy ratio greater than the preset occupancy ratio, the data center with a bandwidth occupancy ratio greater than the preset occupancy ratio is used as the first data center, and the data center with a bandwidth occupancy ratio less than or equal to the preset occupancy ratio is used as the second data center.

[0069] Step S246, if the round-trip time ratio difference of the data packets between the first data center and the second data center is less than or equal to the preset round-trip time ratio difference, determine that the call ratio of the second data center is a preset multiple of the corresponding scheduling benchmark value.

[0070] Step S247, if the round-trip time ratio difference of the data packets between the first data center and the second data center is greater than the preset round-trip time ratio difference, determine the call ratio of the second data center as a preset multiple of the corresponding scheduling benchmark value and the round-trip time ratio difference of the data packet.

[0071] Step S248: Based on the calling ratio of the second data center, determine the corresponding remaining traffic scheduling ratio as the calling ratio of the first data center.

[0072] Exemplarily, when the bandwidth occupancy ratio of the first data center is greater than the preset occupancy ratio, the bandwidth occupancy ratio of the second data center is less than or equal to the preset occupancy ratio, which is set to 90%, and the round-trip time ratio difference of the data packets of the two data centers is less than or equal to the preset round-trip time ratio difference, for example, 30%, for example 90%, 90% and | - | When the bandwidth usage ratio is small or the bandwidth margin is large, the scheduling ratio of the data center with the smaller bandwidth usage ratio or the larger bandwidth margin is prioritized, and the scheduling ratio of the second data center is determined. =2 If the round-trip time ratio difference of the data packets between the two data centers is greater than the preset round-trip time ratio difference, that is 90%, 90% and | - | When , the scheduling ratio of the data center is adjusted according to the difference in the round-trip time ratio of the data packets between the two data centers, for example . In determining After that, the remaining traffic scheduling ratio is used as ,Right now =1- .

[0073] It should be noted that the values ​​of the preset occupancy ratio and the preset round-trip time ratio difference and the specific formula in this embodiment can be adjusted according to actual conditions, and this application does not make specific limitations on this.

[0074] Based on the above embodiments of the present application, in the sixth embodiment of the present application, if there are more than two data centers, the scheduling weight factor of each data center is determined based on the bandwidth occupancy ratio, bandwidth margin and data packet round-trip time of each data center, and then the scheduling ratio of each data center is determined based on the scheduling weight factor of each data center and the preset scheduling reference value.

[0075] Exemplarily, the current bandwidth occupancy ratio, bandwidth margin and data packet round-trip time of the data center are first collected, and the weight allocation principle of the bandwidth occupancy ratio, bandwidth margin and data packet round-trip time is determined according to business needs, resource allocation strategies and other factors. For example, the bandwidth margin can be set to have a higher weight to forward more traffic data to a data center with more available bandwidth resources. For each data center, the individual weight values ​​are calculated separately according to the data of its bandwidth occupancy ratio, bandwidth margin and data packet round-trip time. The calculation method of the individual weight value includes linear function, nonlinear function or exponential function, etc. Afterwards, the individual weight values ​​are subjected to weighted summation or weighted average operations to obtain the scheduling weight factor of each data center.

[0076] Before calculating the scheduling ratio, obtain the preset scheduling benchmark value. According to the scheduling weight factor of each data center and the preset scheduling benchmark value, the scheduling ratio of each data center is determined through certain calculations such as proportional allocation and weighted average, so as to ensure that the traffic data can be fairly and efficiently allocated to each data center according to the actual situation of the data center, and optimize resource utilization.

[0077] Optionally, if a failure is detected in at least one data center, the additional scheduling ratio of each normal data center is determined according to the bandwidth occupancy ratio, bandwidth margin and data packet round trip time of each other normal data center. Then, based on the additional scheduling ratio of each normal data center, the traffic data is forwarded to the normal data centers.

[0078] Exemplarily, after confirming that at least one data center fails, all data centers that are operating normally are identified, and the bandwidth occupancy ratio, bandwidth margin, and data packet round trip time of each normal data center are collected.

[0079] Assign a specific weight value to each of the above indicators, and calculate the additional scheduling ratio based on the weight value. For example, for each data center, calculate the sum of the individual scores of each indicator based on its bandwidth margin, bandwidth occupancy rate, and data packet round-trip time to obtain its comprehensive score. Divide the comprehensive score of each data center by the sum of the comprehensive scores of all data centers, and then multiply it by a preset total amount of additional scheduling to obtain the additional scheduling ratio of each data center, where the total amount of additional scheduling can be determined based on the traffic load of the faulty data center. If there is still some unallocated traffic data after the data traffic is forwarded to the normally operating data center according to the additional scheduling ratio, the unallocated traffic data will be temporarily stored in the edge node traffic proxy module.

[0080] Optionally, when there are only two data centers and one of them cannot provide normal services, all user traffic requests are dispatched to the available data center. Assuming that the entrance of the first data center fails and cannot provide normal services, determine 0, =100%. represents the dispatch ratio of the first data center that fails, Indicates the scheduling ratio of the second data center that is operating normally.

[0081] Optionally, the value of the scheduling ratio of the data center can be set between 5% and 95%.

[0082] It should be noted that when there is no failure in the data center, even if the communication efficiency of a data center is poor, the traffic ratio of the data center should not be set to 0. On the contrary, a part of the traffic should be reserved for the data center. Because if the traffic ratio of a data center is 0, the speed measurement data of the data center cannot be obtained, and the round-trip time of its data packets cannot be evaluated, which affects the traffic scheduling decision. Therefore, even if the communication efficiency of a data center is poor, there may be a small amount of available bandwidth. Reserving a 5% traffic scheduling ratio for the data center can make full use of this part of the bandwidth and avoid resource waste. When the network condition of the data center improves, the reserved part of the traffic can participate in the calculation of the round-trip time of the data packet, helping the traffic ratio of the data center gradually return to normal levels. By reserving a certain traffic ratio for the poorly performing data center, the possible available bandwidth can be fully utilized to avoid complete waste of resources. At the same time, when a data center fails, the edge node traffic proxy module can quickly transfer the traffic to other data centers, reducing the risk of complete failure of the traffic data scheduling system due to a single point failure.

[0083] Based on the above embodiments of the present application, in the seventh embodiment of the present application, the same or similar contents as those in the above embodiments can be referred to the above introduction, and will not be described in detail later. Figure 7 , before step S22, steps S25 to S27 are included: Step S25, obtaining the historical round-trip time of the traffic data from each of the data centers.

[0084] It should be noted that the historical RTT of traffic data refers to the time required for traffic data to be sent from the edge node to the data center and returned from the data center to the edge node, or the round-trip time for traffic data to be distributed from the user end to the data center for processing and then the data request is returned to the user end.

[0085] Step S26, within the first preset time period, respectively calculate the historical round-trip time average of the traffic data in each of the data centers.

[0086] The edge node traffic proxy module periodically collects historical round-trip time data of traffic data from the two data centers. The edge node traffic proxy module sends requests to the two data centers, collects and records the RTT of each request, and calculates the average of all RTTs of the two data centers within a first preset time period.

[0087] Exemplarily, assuming that the first preset time period is 1 minute, the edge node traffic proxy module averages all RTTs collected within 1 minute to obtain the mean round-trip time. For example, the RTTs collected within 1 minute are added together and then divided by the amount of traffic data received within this 1 minute to obtain the mean round-trip time. The mean round-trip time can more accurately reflect the network communication efficiency of the data center and avoid the impact of RTT fluctuations of a single data packet.

[0088] Step S27, within the second preset time period, exponential smoothing calculation is performed on the mean of the historical round-trip time of each of the data centers to obtain the round-trip time of the data packet of each of the data centers.

[0089] It should be noted that the exponential smoothing algorithm can eliminate random fluctuations in the mean round-trip time and reflect the changing trend of the mean round-trip time in the second preset time period.

[0090] Exemplarily, assume that the basic formula of the exponential smoothing algorithm is as follows:

[0091] in, It is the smoothed value at the t-th time point in the second preset time period, that is, the estimated value of the communication time. is the actual observed value at the tth time point, that is, the mean round-trip time. It is the smoothing coefficient, which takes a value between 0 and 1 and is used to control the degree of smoothing. is the smoothed value at the t-1th time point in the second preset time period.

[0092] Before starting exponential smoothing, you can initialize the smoothed value of the first time point in the second time period, such as the actual observed value at the first time point: For each data center, starting from the second time point, the exponential smoothing formula is applied to calculate the smoothed value of each time point until the smoothed values ​​of all time points in the second preset time period are calculated. The smoothed value at the last time point is selected as the communication time estimate of the data center, and the communication time is used as the round-trip time of the data packet of the data center.

[0093] In this embodiment, by calculating the historical round-trip time mean of the data center data traffic and applying an exponential smoothing algorithm, an accurate and flexible method is provided to estimate the data packet round-trip time of the data center, which not only improves the accuracy of the data packet round-trip time estimation, but also helps to optimize the traffic data scheduling decisions and forward the traffic data more efficiently to a data center with better performance.

[0094] Based on the above embodiments of the present application, in the eighth embodiment of the present application, the same or similar contents as those in the above embodiments can be referred to the above introduction, and will not be repeated in the following. On this basis, the traffic scheduling method of the multi-data center system also includes: If the unique token value corresponding to the traffic data exists in the designated data center, the traffic data is forwarded to the designated data center.

[0095] When traffic data is first received, the system checks whether the unique token value associated with the traffic data has a special correspondence with a specific data center. If the unique token value corresponding to the traffic data exists in the specified data center, the system will dispatch the user's traffic request to that specific data center for processing.

[0096] Exemplarily, a special user list can be set up in the unified user center module of the data center, including the tokens of the special users designated by the data center. The edge node traffic proxy module can obtain the special user list of the data center from the parameter control module. When receiving traffic data, it is determined whether the user's token is in the special user list of the data center. If it is in the special user list, the user's request is dispatched to the data center corresponding to the special list; if it is not in the special user list, the traffic data is allocated according to the dispatch ratio of the data center.

[0097] In an optional implementation, the traffic scheduling method for the multi-data center system further includes: forwarding the traffic data to the historically corresponding data center according to the unique token value associated with the traffic data and the historical correspondence between the data centers.

[0098] In order to maintain the continuity of user services in the data center, the edge node traffic proxy module will record the correspondence between the traffic data and the assigned data center after each dispatch of traffic data. When a user has previously initiated a request and the system has processed the user's request at least once, when the user sends a request to the edge node traffic proxy module again, that is, when the edge node traffic proxy module receives traffic data with the same unique token value again, it will check the historical records, find the data center that processed the traffic data last, and dispatch the new request to the same data center, reducing data migration and replication, improving processing efficiency, and maintaining the continuity of user services.

[0099] Optionally, if the edge node traffic proxy module receives traffic data with the same unique token value again and finds that the data center that last processed the traffic data has failed, the traffic data can be scheduled to another data center. After the original data center is restored, the traffic data can be returned to the original data center.

[0100] In this embodiment, by directly scheduling traffic data to a designated data center or a historically corresponding data center, the need for data migration and replication can be reduced. Traffic data is sent to the data center to which they have been connected before, which can also reduce the risk of service interruption due to data center switching, thereby helping to maintain the continuity of user services.

[0101] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the traffic scheduling method of the multi-data center system of the present application. More simple transformations based on this technical concept are all within the scope of protection of the present application.

[0102] The present application provides a traffic scheduling device for a multi-data center system, the traffic scheduling device for the multi-data center system comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the traffic scheduling method for the multi-data center system in the above-mentioned embodiment one.

[0103] Reference below Figure 8 , which shows a schematic diagram of the structure of a traffic scheduling device for a multi-data center system suitable for implementing an embodiment of the present application. The traffic scheduling device for a multi-data center system in an embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptops, digital broadcast receivers, PADs (Portable Application Description: tablet computers), PMPs (Portable Media Player: portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 8 The traffic scheduling device for the multi-data center system shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0104] like Figure 8As shown, the traffic scheduling device of the multi-data center system may include a processing device 1001 (such as a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM: Read Only Memory) 1002 or the program loaded from the storage device 1003 to the random access memory (RAM: Random Access Memory) 1004. In RAM1004, various programs and data required for the operation of the traffic scheduling device of the multi-data center system are also stored. The processing device 1001, ROM1002 and RAM1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems can be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, etc.; an output device 1008 including, for example, a liquid crystal display (LCD: Liquid Crystal Display), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the traffic scheduling device of the multi-data center system to communicate with other devices wirelessly or by wire to exchange data. Although the figure shows the traffic scheduling device of the multi-data center system with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems can be implemented or provided instead.

[0105] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0106] The traffic scheduling device for a multi-data center system provided by the present application adopts the traffic scheduling method for a multi-data center system in the above-mentioned embodiment, which can solve the technical problem that the current multi-data center system architecture cannot schedule traffic requests in real time. Compared with the prior art, the beneficial effects of the traffic scheduling device for a multi-data center system provided by the present application are the same as the beneficial effects of the traffic scheduling method for a multi-data center system provided by the above-mentioned embodiment, and other technical features in the traffic scheduling device for a multi-data center system are the same as the features disclosed in the method of the previous embodiment, which will not be described in detail here.

[0107] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0108] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0109] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer programs) stored thereon, the computer-readable program instructions being used to execute the traffic scheduling method for the multi-data center system in the above-mentioned embodiment.

[0110] The computer-readable storage medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM: Random Access Memory), a read-only memory (ROM: Read Only Memory), an erasable programmable read-only memory (EPROM: Erasable Programmable Read Only Memory or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM: CD-Read Only Memory), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency: Radio Frequency), etc., or any suitable combination of the above.

[0111] The computer-readable storage medium may be included in the traffic scheduling device of the multi-data center system; or may exist independently without being assembled into the traffic scheduling device of the multi-data center system.

[0112] The computer-readable storage medium carries one or more programs. When the one or more programs are executed by the traffic scheduling device of the multi-data center system, the traffic scheduling device of the multi-data center system can be written in one or more programming languages ​​or a combination thereof to perform the computer program code of the operation of the present application. The programming language includes an object-oriented programming language, such as Java, Smalltalk, C++, and a conventional procedural programming language, such as "C" language or a similar programming language. The program code can be executed completely on the user's computer, partially on the user's computer, or as an independent software package, partially on the user's computer and partially on a remote computer, or completely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (for example, using an Internet service provider to connect through the Internet).

[0113] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0114] The modules involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.

[0115] The readable storage medium provided in the present application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned traffic scheduling method for the multi-data center system, and can solve the technical problem that the multi-data center system architecture cannot schedule traffic requests in real time. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in the present application are the same as the beneficial effects of the traffic scheduling method for the multi-data center system provided in the above-mentioned embodiment, and will not be repeated here.

[0116] The present application also provides a computer program product, including a computer program, which implements the steps of the traffic scheduling method for the multi-data center system as described above when the computer program is executed by a processor.

[0117] The computer program product provided by the present application can solve the technical problem that the current multi-data center system architecture cannot schedule traffic requests in real time. Compared with the prior art, the beneficial effects of the computer program product provided by the present application are the same as the beneficial effects of the traffic scheduling method of the multi-data center system provided by the above embodiment, which will not be repeated here.

[0118] The above descriptions are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect applications in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A traffic scheduling method for a multi-data center system, characterized in that: Applied to edge nodes, the traffic scheduling method of the multi-data center system includes: When the edge node receives the flow data from the user side, obtaining a unique token value corresponding to the flow data; If the unique token value does not specify the data center, determine the scheduling ratio of each data center based on the bandwidth occupancy ratio, bandwidth margin and data packet round trip time of each data center; Based on the scheduling ratio of the data center, a target data center is determined, and the traffic data is forwarded to the target data center.

2. The traffic scheduling method for a multi-data center system according to claim 1, characterized in that: The step of determining the scheduling ratio of each data center based on the bandwidth occupancy ratio, bandwidth margin and data packet round trip time of each data center includes: Obtaining the bandwidth occupancy ratio and bandwidth margin of each of the data centers within a preset time interval; Determine the round trip time ratio difference of the data packets between the data centers based on the round trip time of the data packets of the data centers; Obtaining a scheduling benchmark value for each of the data centers; The scheduling ratio of each of the data centers is determined based on the bandwidth occupancy ratio, the bandwidth margin, the scheduling reference value, and the data packet round-trip time ratio difference of each of the data centers.

3. The traffic scheduling method for a multi-data center system according to claim 2, characterized in that: The step of determining the scheduling ratio of each data center based on the bandwidth occupancy ratio, the bandwidth margin, the scheduling reference value, and the data packet round trip time ratio difference of each data center includes: If the bandwidth occupancy ratio of each of the data centers is less than or equal to the preset occupancy ratio, and the data packet round-trip time ratio difference is less than or equal to the preset round-trip time ratio difference, determining the scheduling ratio of each of the data centers according to the scheduling reference value; If the bandwidth occupancy ratio of each of the data centers is less than or equal to the preset occupancy ratio, and the data packet round-trip time ratio difference is greater than the preset round-trip time ratio difference, the scheduling ratio of each of the data centers is determined according to the scheduling reference value and the data packet round-trip time ratio difference.

4. The traffic scheduling method for a multi-data center system according to claim 2, characterized in that: The step of determining the scheduling ratio of each data center based on the bandwidth occupancy ratio, the bandwidth margin, the scheduling reference value, and the data packet round trip time ratio difference of each data center also includes: If the bandwidth occupancy ratio of each of the data centers is greater than the preset occupancy ratio, and the round-trip time ratio difference of the data packets is less than or equal to the preset round-trip time ratio difference, determining the scheduling ratio of each of the data centers according to the scheduling reference value and the bandwidth margin; If the bandwidth occupancy ratio of each of the data centers is greater than the preset occupancy ratio, and the data packet round-trip time ratio difference is greater than the preset round-trip time ratio difference, the scheduling ratio of each of the data centers is determined according to the scheduling reference value, the bandwidth margin and the data packet round-trip time ratio difference.

5. The traffic scheduling method for a multi-data center system according to claim 2, characterized in that: The step of determining the scheduling ratio of each data center based on the bandwidth occupancy ratio, the bandwidth margin, the scheduling reference value, and the data packet round trip time ratio difference of each data center also includes: If there is a data center with a bandwidth occupancy ratio greater than the preset occupancy ratio, the data center with a bandwidth occupancy ratio greater than the preset occupancy ratio is used as the first data center, and the data center with a bandwidth occupancy ratio less than or equal to the preset occupancy ratio is used as the second data center; If the round-trip time ratio difference of the data packets between the first data center and the second data center is less than or equal to the preset round-trip time ratio difference, determining that the call ratio of the second data center is a preset multiple of the corresponding scheduling reference value; If the round-trip time ratio difference of the data packets between the first data center and the second data center is greater than the preset round-trip time ratio difference, determining the call ratio of the second data center to be a preset multiple of the corresponding scheduling reference value and the round-trip time ratio difference of the data packets; Based on the calling ratio of the second data center, the corresponding remaining traffic scheduling ratio is determined to be the calling ratio of the first data center.

6. The traffic scheduling method for a multi-data center system according to claim 2, characterized in that: The step of determining the round trip time ratio difference of data packets between the data centers based on the round trip time of the data packets of the data centers includes: Obtaining historical round-trip time of traffic data from each of the data centers; In a first preset time period, respectively calculating the average of the historical round-trip time of the traffic data in each of the data centers; In the second preset time period, exponential smoothing calculation is performed on the mean of the historical round-trip time of each data center to obtain the round-trip time of the data packet of each data center.

7. The traffic scheduling method for a multi-data center system according to claim 1, characterized in that: The method further comprises: If the unique token value corresponding to the traffic data exists in the designated data center, forwarding the traffic data to the designated data center; Alternatively, the traffic data is forwarded to the historically corresponding data center according to the unique token value associated with the traffic data and the historical corresponding relationship between the data centers.

8. A traffic scheduling device for a multi-data center system, characterized in that: The device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the traffic scheduling method for a multi-data center system according to any one of claims 1 to 7.

9. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the traffic scheduling method for a multi-data center system as described in any one of claims 1 to 7 are implemented.

10. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the steps of the traffic scheduling method for a multi-data center system according to any one of claims 1 to 8 are implemented.

Images