A data transmission method and related equipment
By introducing cache devices into the cloud service system and splitting the data transmission path, the network congestion problem caused by server path collision is solved, efficient data transmission is achieved, and the computing power requirements of the arbitrator are reduced.
Patent Information
- Application Number
- CN202110707688.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-06-24
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2041-06-24
AI Technical Summary
In the cloud service system, the transmission paths between servers may collide, resulting in network congestion. The computing power of the existing arbitrators is limited, and it is impossible to effectively solve the network congestion problem.
By setting up cache devices between servers and splitting the data transmission path, the arbitrator only needs to plan the path between the cache devices and their respective servers, share the scheduling pressure, and avoid high computing power requirements.
It effectively solves the problem of network congestion, reduces the requirements for arbitrator computing power, and improves data transmission efficiency.
Smart Images

Figure CN115529276B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of cloud technology, and in particular to a data transmission method and related equipment. Background Art
[0002] With the rapid development of cloud technology, the scale of cloud service systems is getting larger and larger. Figure 1 It is a structural diagram of a cloud service system in the related technology, such as Figure 1 As shown, the cloud service system has a three-layer architecture, with the bottom layer being widely distributed physical servers, the middle layer being top of rack (TOR) switches, and the top layer being core switches. Data transmission between different servers can be achieved through two layers of switches.
[0003] Due to the large scale of cloud service systems, the transmission paths between servers may collide (for example, the path between server 1 and server 2, and the path between server 3 and server 4 have overlapping parts, and the overlapping parts of the paths can only process the data between server 1 and server 2 first, and then process the data between server 3 and server 4), resulting in network congestion and seriously affecting data transmission efficiency. In order to solve this problem, a cloud service system is usually equipped with an arbitrator that manages all servers. The arbitrator can implement path scheduling, that is, the arbitrator can plan paths for multiple pairs of servers that will not collide, thereby avoiding network congestion.
[0004] However, the computing power of the arbitrator is often limited and cannot complete the path scheduling of all servers in a short time, so it cannot effectively solve the problem of network congestion. Summary of the invention
[0005] The embodiments of the present application provide a data transmission method and related devices thereof, which can effectively solve the problem of network congestion without placing excessively high requirements on the computing power of the arbitrator.
[0006] A first aspect of an embodiment of the present application provides a data transmission method, the method comprising:
[0007] When the first server needs to send the first data to the second server, the first arbitrator managing the first server can plan a first path for the first server, where the first path starts from the first server and ends at the first cache device. Then, the first server can send the first data to the first cache device along the first path for temporary storage.
[0008] After receiving the first data from the first server, the first cache device can determine that the first data is data to be sent to the second server, so the first cache device can send a first request to the second arbitrator that manages the second server. The first request is used to indicate that the first cache device needs to send the first data to the second server.
[0009] After receiving the first request, the second arbitrator can determine that the first cache device needs to send the first data to the second server, so the second arbitrator can plan a second path for the first cache device. The second path starts at the first cache device and ends at the second server, and notifies the first cache device through the first response.
[0010] After the first cache device receives the first response from the second arbitrator, since the first response is used to indicate the second path, the first cache device can send the first data to the second server along the second path. At this point, the data transmission between the first server and the second server is completed.
[0011] From the above method, it can be seen that: a first cache device is set between the first server and the second server as a forwarding point for temporarily storing data. Then, the data transmission path between the first server and the second server can be divided into two parts. The first path between the first server and the first cache device can be planned by the first arbitrator who manages the first server, and the second path between the first cache device and the second server can be planned by the second arbitrator who manages the second server. It can be seen that the first server as the data sending end is provided with a first arbitrator, and the second server as the data receiving end is provided with a second arbitrator. The two arbitrators can jointly complete the path scheduling between the first server and the second server and share the scheduling pressure. Therefore, there is no need to make excessive requirements on the computing power of the arbitrator, and the problem of network congestion can be effectively solved.
[0012] In a possible implementation, the first server and the first arbitrator are located in the first logical area, and the second server and the second arbitrator are located in the second logical area. In the aforementioned implementation, the entire cloud service system can be divided into multiple logical areas, each of which has an arbitrator deployed. Each arbitrator can manage all servers in its logical area. For example, the first arbitrator in the first logical area can manage all servers in the first logical area (including the first server), and the second arbitrator in the second logical area can manage all servers in the second logical area (including the second server).
[0013] In a possible implementation, the first logical area and the second logical area are regions, availability zones, data centers, or chassis.
[0014] In a possible implementation, before the first cache device sends the first request to the second arbitrator, the method further includes: the first cache device determines the amount of data to be sent to the multiple logical areas and the amount of data to be sent to the second logical area based on the first data, and the multiple logical areas include the first logical area and the second logical area; the first cache device sends the first request to the second arbitrator including: if the amount of data to be sent to the multiple logical areas is less than or equal to the preset data amount threshold, and the amount of data to be sent to the second logical area is within the preset value range, the first cache device sends the first request to the second arbitrator. In the aforementioned implementation, if the amount of data to be sent to the multiple logical areas is less than or equal to the preset data amount threshold, and the amount of data to be sent to the second logical area is within the preset value range, the first cache device can determine that the cache of the first cache device has not overflowed, and the port length of the second logical area is normal, so the first cache device sends the first request to the second arbitrator, so that the second arbitrator plans the second path for the first cache device based on the first request.
[0015] In a possible implementation, the method further includes: if the amount of data to be sent to multiple logical areas is greater than a data amount threshold, the first cache device sends a second request to the arbitrators in the multiple logical areas, and the second request is used to instruct the first cache device to refuse to send data to the multiple logical areas. In the aforementioned implementation, if the amount of data to be sent to multiple logical areas is greater than a preset data amount threshold, the first cache device can determine that there is a risk of overflow in the cache of the first cache device, so the first cache device sends a second request to the arbitrators in the multiple logical areas. After the arbitrators of the multiple logical areas receive the second request, each arbitrator will prohibit the server it manages from sending data to each logical area through the first cache device (it is considered that the first cache device is unavailable to all logical areas).
[0016] In a possible implementation, the method further includes: if the amount of data to be sent to the multiple logical areas is less than or equal to the data amount threshold, and the amount of data to be sent to the second logical area is outside the value range, the first cache device sends a third request to the arbitrators in the multiple logical areas, and the third request is used to instruct the first cache device to refuse to send data to the second logical area. In the aforementioned implementation, if the amount of data to be sent to the multiple logical areas is less than or equal to the preset data amount threshold, and the amount of data to be sent to the second logical area is outside the preset value range, the first cache device can determine that the port captain of the second logical area is abnormal, and then send the third request to the arbitrators in the multiple logical areas. After the arbitrators of the multiple logical areas receive the third request, each arbitrator will prohibit the server it manages from sending data to the second logical area through the first cache device (considering that the first cache device is unavailable to the second logical area).
[0017] In a possible implementation, the preset data volume threshold can be determined based on the time when the first cache device communicates with the first arbitrator, the time when the first arbitrator processes information, the time when the first arbitrator communicates with the first server, and the time when the first server communicates with the first cache device. In this way, the first cache device can compare the amount of data to be sent to the multiple logical areas with the data volume threshold, thereby determining whether the cache of the first cache device overflows.
[0018] In a possible implementation, the left endpoint of the preset value range is determined according to the time when the first cache device communicates with the first arbitrator, the time when the first arbitrator processes information, the time when the first arbitrator communicates with the first server, and the time when the first server communicates with the first cache device, and the right endpoint of the preset value range is a preset value. In this way, the first cache device can determine whether the amount of data to be sent to the second logical area is within the preset value range, thereby determining whether the port captain of the second logical area is normal.
[0019] In a possible implementation, the first data carries the identifier of the second server, and the first cache device sends a first request to the second arbitrator that manages the second server, including: the first cache device determines the second arbitrator that manages the second server according to the identifier of the second server; the first cache device sends the first request to the second arbitrator. In the aforementioned implementation, the first data sent by the first server to the first cache device carries the identifier of the second server, so after receiving the first data, the first cache device can determine that the first data needs to be forwarded to the second server based on the identifier of the second server, so it can determine the second arbitrator that manages the second server, and send a first request to the second arbitrator, so that the second arbitrator plans a second path for the first cache device based on the first request.
[0020] In a possible implementation manner, the first cache device is a switch or a server.
[0021] In a possible implementation manner, the first arbitrator and the second arbitrator are servers, offload cards, or heterogeneous cards.
[0022] A second aspect of an embodiment of the present application provides a data transmission method, the method comprising:
[0023] When the first server needs to send the first data to the second server, the first server may send a fourth request to the first arbitrator that manages the first server, where the fourth request is used to instruct sending the first data to the second server, and the first arbitrator manages the first server.
[0024] After the first arbitrator receives the fourth request, it can be determined based on the fourth request that the first server needs to send the first data to the second server. Therefore, the first arbitrator can determine the first cache device from multiple cache devices and generate a first path. The first path starts at the first server and ends at the first cache device.
[0025] After obtaining the first path, the first arbitrator sends a fourth response to the first server, and the fourth response is used to instruct the first server to send the first data to be sent to the second server to the first cache device along the first path. Then, after receiving the fourth response, the first server can send the first data to the first cache device along the first path indicated by the fourth response for temporary storage.
[0026] Subsequently, the first cache device sends a request to the second arbitrator that manages the second server, so that the second arbitrator plans a second path for the first cache device, the second path starting from the first cache device and ending at the second server. Therefore, the first cache device can send the first data to the second server along the second path.
[0027] From the above method, it can be seen that: a first cache device is set between the first server and the second server as a forwarding point for temporarily storing data. Then, the data transmission path between the first server and the second server can be divided into two parts. The first path between the first server and the first cache device can be planned by the first arbitrator who manages the first server, and the second path between the first cache device and the second server can be planned by the second arbitrator who manages the second server. It can be seen that the first server as the data sending end is provided with a first arbitrator, and the second server as the data receiving end is provided with a second arbitrator. The two arbitrators can jointly complete the path scheduling between the first server and the second server and share the scheduling pressure. Therefore, there is no need to make excessive requirements on the computing power of the arbitrator, and the problem of network congestion can be effectively solved.
[0028] In a possible implementation manner, the first server and the first arbitrator are located in a first logical area, and the second server and a second arbitrator that manages the second server are located in a second logical area.
[0029] In a possible implementation, the first logical area and the second logical area are regions, availability zones, data centers, or chassis.
[0030] In a possible implementation, before the first arbitrator determines the first cache device from the plurality of cache devices according to the fourth request, the method further includes: the first arbitrator receives a fifth request from the second cache device, the fifth request is used to instruct the second cache device to refuse to send data to the plurality of logical areas, the plurality of logical areas including the first logical area and the second logical area; the first arbitrator determines the first cache device from the plurality of cache devices according to the fourth request, including: the first arbitrator removes the second cache device from the plurality of cache devices according to the fourth request and the fifth request, and selects the first cache device from the remaining cache devices. In the aforementioned implementation, the first arbitrator may centrally process the plurality of requests received in the current time period. In the current time period, if the first arbitrator receives a fifth request sent by the second cache device to the first arbitrator in addition to the fourth request, and the fifth request is used to instruct the second cache device to refuse to send data to the plurality of logical areas (for example, all logical areas of the cloud service system), the first arbitrator may determine that the second cache device is unavailable for the plurality of logical areas based on the request, then the first arbitrator may remove the second cache device from all cache devices, and randomly select a cache device from the remaining cache devices as the first cache device.
[0031] In a possible implementation, before the first arbitrator determines the first cache device from the plurality of cache devices according to the fourth request, the method further includes: the first arbitrator receives a sixth request from the second cache device, the sixth request being used to instruct the second cache device to refuse to send data to the second logical area; the first arbitrator determines the first cache device from the plurality of cache devices according to the fourth request, including: the first arbitrator removes the second cache device from the plurality of cache devices according to the fourth request and the sixth request, and selects the first cache device from the remaining cache devices. In the aforementioned implementation, the first arbitrator may centrally process the plurality of requests received in the current time period. In the current time period, if the first arbitrator receives a sixth request sent by the second cache device to the first arbitrator in addition to the fourth request, and the sixth request is used to instruct the second cache device to refuse to send data to the second logical area, the first arbitrator may determine that the second cache device is unavailable for the second logical area based on the request, then the first arbitrator may remove the second cache device from all cache devices, and randomly select a cache device from the remaining cache devices as the first cache device.
[0032] In a possible implementation, after the first arbitrator sends the fourth response to the first server, the method further includes: the first arbitrator receives a second request from the first cache device, the second request is used to instruct the first cache device to refuse to send data to multiple logical areas; the first arbitrator removes the first cache device from the multiple cache devices according to the second request. In the aforementioned implementation, after the first arbitrator receives the second request from the first cache device (the second request is used to instruct the first cache device to refuse to send data to multiple logical areas), the first arbitrator usually enters the next time period at this time. When the first arbitrator centrally processes multiple requests in the next time period, it can remove the first cache device from the multiple cache devices according to the second request.
[0033] In a possible implementation, after the first arbitrator sends the fourth response to the first server, the method further includes: the first arbitrator receives a third request from the first cache device, the third request is used to instruct the first cache device to refuse to send data to the second logical area; the first arbitrator removes the first cache device from the multiple cache devices according to the third request. In the aforementioned implementation, after the first arbitrator receives the third request from the first cache device (the second request is used to instruct the first cache device to refuse to send data to the second logical area), the first arbitrator usually enters the next time period at this time. When the first arbitrator centrally processes multiple requests in the next time period, it can remove the first cache device from the multiple cache devices according to the third request.
[0034] In a possible implementation, before the first arbitrator determines the first cache device from multiple cache devices according to the fourth request and generates the first path, the method further includes: the first arbitrator receives the seventh request from the first server, the seventh request is used to indicate that the second data is sent to the third server, and the first arbitrator and the third server are located in the first logical area; the first arbitrator selects one of the requests from the fourth request and the seventh request for priority processing according to the preset request priority. In the above implementation, in the current time period, if the first arbitrator receives the seventh request sent by the first server to the first arbitrator in addition to the fourth request, the seventh request can be used to indicate that the first server needs to send the second data to the third server in the first logical area. Based on the fourth request and the seventh request, the first arbitrator can determine that the first server needs to send not only the first data to the second server, but also the second data to the third server, that is, the first server needs to realize data transmission across logical areas and data transmission within the logical area. Then, the first arbitrator can select one of the requests from the fourth request and the seventh request for priority processing according to the preset request priority.
[0035] In a possible implementation manner, the first data carries an identifier of the second server.
[0036] In a possible implementation manner, the first cache device and the second cache device are switches or servers.
[0037] In a possible implementation manner, the first arbitrator and the second arbitrator are servers, offload cards, or heterogeneous cards.
[0038] In a possible implementation, the priority of the seventh request is usually higher than the priority of the fourth request.
[0039] A third aspect of the embodiments of the present application provides a data transmission method, the method comprising:
[0040] When the first server needs to send the first data to the second server, the first arbitrator managing the first server can plan a first path for the first server, where the first path starts from the first server and ends at the first cache device. Then, the first server can send the first data to the first cache device along the first path for temporary storage.
[0041] After receiving the first data from the first server, the first cache device can determine that the first data is data to be sent to the second server, so the first cache device can send a first request to the second arbitrator that manages the second server. The first request is used to indicate that the first cache device needs to send the first data to the second server.
[0042] After receiving the first request, the second arbitrator can determine that the first cache device needs to send the first data to the second server. Therefore, the second arbitrator can plan a second path for the first cache device. The second path starts at the first cache device and ends at the second server. The first cache device is notified through the first response, so that the first cache device sends the first data to the second server along the second path.
[0043] From the above method, it can be seen that: a first cache device is set between the first server and the second server as a forwarding point for temporarily storing data. Then, the data transmission path between the first server and the second server can be divided into two parts. The first path between the first server and the first cache device can be planned by the first arbitrator who manages the first server, and the second path between the first cache device and the second server can be planned by the second arbitrator who manages the second server. It can be seen that the first server as the data sending end is provided with a first arbitrator, and the second server as the data receiving end is provided with a second arbitrator. The two arbitrators can jointly complete the path scheduling between the first server and the second server and share the scheduling pressure. Therefore, there is no need to make excessive requirements on the computing power of the arbitrator, and the problem of network congestion can be effectively solved.
[0044] In a possible implementation manner, the second server and the second arbitrator are located in the second logical area.
[0045] In a possible implementation, the second logical area is a region, an availability zone, a data center, or a chassis.
[0046] In a possible implementation manner, before the second arbitrator generates the second path according to the first request, the method further includes:
[0047] The second arbitrator receives the eighth request from the fourth server, the eighth request is used to indicate that the third data is sent to the second server, and the second arbitrator and the fourth server are located in the second logical area; the second arbitrator selects one of the requests from the first request and the eighth request for priority processing according to the preset request priority. In the above implementation, during the time period when the second arbitrator receives the first request, if the second arbitrator also receives the eighth request sent by the fourth server in the second logical area, the eighth request is used to indicate that the fourth server needs to send the third data to the second server. Based on the first request and the eighth request, the second arbitrator can determine that not only the first cache device needs to send the first data to the second server, but also the fourth server needs to send the third data to the second server, that is, the second server needs to realize both cross-logical area data transmission and intra-logical area data transmission. Then, the second arbitrator can select one of the requests from the first request and the eighth request for priority processing according to the preset request priority.
[0048] In a possible implementation manner, the first cache device is a switch or a server.
[0049] In a possible implementation manner, the second arbitrator is a server, an offload card, or a heterogeneous card.
[0050] In a possible implementation, the priority of the eighth request is usually higher than the priority of the first request.
[0051] A fourth aspect of an embodiment of the present application provides a cache device, which serves as a first cache device. The first cache device includes: a receiving module, which is used to receive first data to be sent to a second server from a first server on a first path, and the first path is determined by a first arbitrator that manages the first server; a sending module, which is used to send a first request to a second arbitrator that manages the second server, and the first request is used to indicate that the first data is sent to the second server; the receiving module is used to receive a first response from the second arbitrator, and the first response is used to indicate a second path, and the first response is obtained by the second arbitrator processing the first request; the sending module is used to send the first data to the second server along the second path indicated by the first response.
[0052] In a possible implementation manner, the first server and the first arbitrator are located in a first logical area, and the second server and the second arbitrator are located in a second logical area.
[0053] In a possible implementation, the first logical area and the second logical area are regions, availability zones, data centers, or chassis.
[0054] In a possible implementation, the first cache device also includes: a processing module, used to determine the amount of data to be sent to multiple logical areas and the amount of data to be sent to the second logical area based on the first data, the multiple logical areas including the first logical area and the second logical area; a sending module, used to send a first request to the second arbitrator if the amount of data to be sent to the multiple logical areas is less than or equal to a preset data amount threshold, and the amount of data to be sent to the second logical area is within a preset value range.
[0055] In a possible implementation, the sending module is further used to send a second request to the arbitrators in the multiple logical areas if the amount of data to be sent to the multiple logical areas is greater than the data amount threshold, and the second request is used to instruct the first cache device to refuse to send data to the multiple logical areas.
[0056] In one possible implementation, the sending module is further used to send a third request to the arbitrator in the multiple logical areas if the amount of data to be sent to the multiple logical areas is less than or equal to the data amount threshold, and the amount of data to be sent to the second logical area is outside the value range, and the third request is used to instruct the first cache device to refuse to send data to the second logical area.
[0057] In a possible implementation, the data volume threshold is determined according to the time when the first cache device communicates with the first arbitrator, the time when the first arbitrator processes information, the time when the first arbitrator communicates with the first server, and the time when the first server communicates with the first cache device.
[0058] In a possible implementation, the left endpoint of the value range is determined according to the time when the first cache device communicates with the first arbitrator, the time when the first arbitrator processes information, the time when the first arbitrator communicates with the first server, and the time when the first server communicates with the first cache device.
[0059] In a possible implementation manner, the processing module is used to determine a second arbitrator that manages the second server according to an identifier of the second server; and the sending module is used to send the first request to the second arbitrator.
[0060] In a possible implementation manner, the first cache device is a switch or a server.
[0061] In a possible implementation manner, the first arbitrator and the second arbitrator are servers, offload cards, or heterogeneous cards.
[0062] A fifth aspect of an embodiment of the present application provides an arbitrator, which serves as a first arbitrator. The first arbitrator includes: a receiving module, used to receive a fourth request from a first server, the fourth request is used to instruct to send first data to a second server, and the first arbitrator manages the first server; a processing module, used to determine a first cache device from a plurality of cache devices according to the fourth request, and generate a first path; a sending module, used to send a fourth response to the first server, the fourth response is used to instruct the first server to send the first data to be sent to the second server to the first cache device along the first path.
[0063] In a possible implementation manner, the first server and the first arbitrator are located in a first logical area, and the second server and a second arbitrator that manages the second server are located in a second logical area.
[0064] In a possible implementation, the first logical area and the second logical area are regions, availability zones, data centers, or chassis.
[0065] In one possible implementation, the receiving module is further used to receive a fifth request from the second cache device, where the fifth request is used to instruct the second cache device to refuse to send data to multiple logical areas, where the multiple logical areas include a first logical area and a second logical area; and the processing module is used to remove the second cache device from the multiple cache devices based on the fourth request and the fifth request, and select the first cache device from the remaining cache devices.
[0066] In one possible implementation, the receiving module is further used to receive a sixth request from the second cache device, where the sixth request is used to instruct the second cache device to refuse to send data to the second logical area; and the processing module is used to remove the second cache device from the multiple cache devices based on the fourth request and the sixth request, and select the first cache device from the remaining cache devices.
[0067] In a possible implementation, the receiving module is further used to receive a second request from the first cache device, where the second request is used to instruct the first cache device to refuse to send data to multiple logical areas; the processing module is further used to remove the first cache device from the multiple cache devices based on the second request.
[0068] In a possible implementation, the receiving module is further used to receive a third request from the first cache device, where the third request is used to instruct the first cache device to refuse to send data to the second logical area; the processing module is further used to remove the first cache device from the multiple cache devices based on the third request.
[0069] In one possible implementation, the receiving module is further used to receive a seventh request from the first server, the seventh request is used to indicate sending second data to the third server, and the first arbitrator and the third server are located in the first logical area; the processing module is further used to select one of the fourth request and the seventh request for priority processing according to a preset request priority.
[0070] In a possible implementation manner, the first data carries an identifier of the second server.
[0071] In a possible implementation manner, the first cache device and the second cache device are switches or servers.
[0072] In a possible implementation manner, the first arbitrator and the second arbitrator are servers, offload cards, or heterogeneous cards.
[0073] In a possible implementation, the priority of the seventh request is usually higher than the priority of the fourth request.
[0074] The sixth aspect of an embodiment of the present application provides an arbitrator, which serves as a second arbitrator, and the second arbitrator includes: a receiving module, used to receive a first request from a first cache device, the first request is used to indicate sending first data to a second server; a processing module, used to generate a second path according to the first request; a sending module, used to send a first response to the first cache device, the first response is used to indicate the first cache device to send the first data to the second server along the second path.
[0075] In a possible implementation manner, the second server and the second arbitrator are located in the second logical area.
[0076] In a possible implementation, the second logical area is a region, an availability zone, a data center, or a chassis.
[0077] In one possible implementation, the receiving module is further used to receive an eighth request from a fourth server, where the eighth request is used to instruct sending third data to the second server, and the second arbitrator and the fourth server are located in a second logical area; the processing module is further used to select one of the first request and the eighth request for priority processing based on a preset request priority.
[0078] In a possible implementation manner, the first cache device is a switch or a server.
[0079] In a possible implementation manner, the second arbitrator is a server, an offload card, or a heterogeneous card.
[0080] In a possible implementation, the priority of the eighth request is usually higher than the priority of the first request.
[0081] A seventh aspect of an embodiment of the present application provides a cache device, which serves as a first cache device. The first cache device includes a memory and a processor; the memory stores code, and the processor is configured to execute the code. When the code is executed, the first cache device executes the method described in the first aspect or any possible implementation method of the first aspect.
[0082] An eighth aspect of an embodiment of the present application provides an arbitrator, which serves as a first arbitrator. The first arbitrator includes a memory and a processor; the memory stores code, and the processor is configured to execute the code. When the code is executed, the first arbitrator executes the method described in the second aspect or any possible implementation method of the second aspect.
[0083] A ninth aspect of an embodiment of the present application provides an arbitrator, which serves as a second arbitrator, and the second arbitrator includes a memory and a processor; the memory stores code, and the processor is configured to execute the code. When the code is executed, the second arbitrator executes the method described in the third aspect or any possible implementation method of the third aspect.
[0084] The tenth aspect of the embodiment of the present application provides a cloud service system, which includes: a first server, a second server, a first cache device as described in the seventh aspect, a first arbitrator as described in the eighth aspect, and a second arbitrator as described in the ninth aspect; the first end of the first server is connected to the first arbitrator, and the second end of the first server is connected to the first end of the first cache device; the first end of the second server is connected to the second arbitrator, and the second end of the second server is connected to the second end of the first cache device.
[0085] In an eleventh aspect of an embodiment of the present application, a computer storage medium is provided, which stores one or more instructions. When the instructions are executed by one or more computers, the one or more computers implement the method described in the first aspect, any possible implementation method of the first aspect, the second aspect, any possible implementation method of the second aspect, the third aspect, or any possible implementation method of the third aspect.
[0086] The twelfth aspect of the embodiments of the present application provides a computer program product, which stores instructions. When the instructions are executed by a computer, the computer implements the method described in the first aspect, any possible implementation method of the first aspect, the second aspect, any possible implementation method of the second aspect, the third aspect, or any possible implementation method of the third aspect.
[0087] In an embodiment of the present application, when the first server needs to send the first data to the second server, the first arbitrator that manages the first server can plan a first path for the first server so that the first server sends the first data to the first cache device along the first path for caching. Then, the first cache device can send a first request to the second arbitrator that manages the second server to notify the first cache device that the first data needs to be sent to the second server, so the second arbitrator can plan a second path for the first cache device so that the first cache device sends the first data to the second server along the second path. In the above process, a first cache device is provided between the first server and the second server as a forwarding point for temporarily storing data. Then, the data transmission path between the first server and the second server can be divided into two parts. The first path between the first server and the first cache device can be planned by the first arbitrator that manages the first server, and the second path between the first cache device and the second server can be planned by the second arbitrator that manages the second server. It can be seen that a first arbitrator is provided at the first server as the data sending end, and a second arbitrator is provided at the second server as the data receiving end. The two arbitrators can jointly complete the path scheduling between the first server and the second server and share the scheduling pressure. Therefore, there is no need to make excessively high requirements on the computing power of the arbitrator, and the problem of network congestion can be effectively solved. BRIEF DESCRIPTION OF THE DRAWINGS
[0088] Figure 1 A schematic diagram of a cloud service system in the related art;
[0089] Figure 2 A schematic diagram of the structure of a cloud service system provided in an embodiment of the present application;
[0090] Figure 3 A schematic diagram of a flow chart of a data transmission method provided in an embodiment of the present application;
[0091] Figure 4 A schematic diagram of the headroom time provided for an embodiment of the present application;
[0092] Figure 5 Another structural diagram of the cloud service system provided in the embodiment of the present application;
[0093] Figure 6 A schematic diagram of the structure of a cache device provided in an embodiment of the present application;
[0094] Figure 7 A schematic diagram of the structure of an arbitrator provided in an embodiment of the present application;
[0095] Figure 8 Another structural schematic diagram of the arbitrator provided in the embodiment of the present application;
[0096] Fig. 9 Another schematic diagram of the structure of the cache device provided in the embodiment of the present application;
[0097] Fig.10 Another structural schematic diagram of the arbitrator provided in the embodiment of the present application;
[0098] Fig.11 Another structural diagram of the arbitrator provided in an embodiment of the present application. DETAILED DESCRIPTION
[0099] The embodiments of the present application provide a data transmission method and related devices thereof, which can effectively solve the problem of network congestion without placing excessively high requirements on the computing power of the arbitrator.
[0100] The terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and need not be used to describe a specific order or sequential order. It should be understood that the terms used in this way can be interchangeable under appropriate circumstances, which is only to describe the distinction mode adopted by the objects of the same attributes when describing in the embodiments of the present application. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, so that the process, method, system, product or equipment comprising a series of units need not be limited to those units, but may include other units that are not clearly listed or inherent to these processes, methods, products or equipment.
[0101] With the rapid development of cloud technology, the scale of cloud service systems is getting larger and larger. Figure 1 It is a structural diagram of a cloud service system in the related technology, such as Figure 1 As shown, the cloud service system has a three-layer architecture, with the bottom layer being widely distributed physical servers, the middle layer being top of rack (TOR) switches, and the top layer being core switches. Data transmission between different servers can be achieved through two layers of switches.
[0102] Due to the large scale of the cloud service system, the transmission paths between servers may collide, causing network congestion, which seriously affects the data transmission efficiency. For example, the path between server 1 and server 2 is: server 1→TOR switch 1→core switch 1→TOR switch 2→server 2. The path between server 3 and server 4 is: server 3→TOR switch 1→core switch 1→TOR switch 2→server 4. It can be seen that there is an overlap (collision) between the two paths, namely the path of TOR switch 1→core switch 1→TOR switch 2. Then, when TOR switch 1 receives data sent from server 1 to server 2 and data sent from server 3 to server 4, TOR switch 1 can only send data from server 1 to core switch 1 first, and then send data from server 3 (or send data from server 3 to core switch 1 first, and then send data from server 1), resulting in low data transmission efficiency.
[0103] In order to solve the network congestion, the cloud service system is usually equipped with an arbitrator that manages all servers. The arbitrator can implement path scheduling, that is, the arbitrator can plan paths for multiple pairs of servers that will not collide, thereby avoiding network congestion. As in the above example, the arbitrator can keep the path between server 3 and server 4 unchanged, and re-plan the path between server 3 and server 4 as follows: server 3→TOR switch 1→core switch 2→TOR switch 2→server 4. It can be seen that there is no overlap between the two paths (no collision), so there will be no network congestion when transmitting data between server 1 and server 2 and between server 3 and server 4.
[0104] However, the computing power of the arbitrator is often limited and cannot complete the path scheduling of all servers in a short time, so it cannot effectively solve the problem of network congestion.
[0105] In order to solve the above problems, the present application provides a data transmission method, which can be applied to Figure 2 The cloud service system shown ( Figure 2 A schematic diagram of the structure of the cloud service system provided in the embodiment of the present application). The cloud service system is first introduced below. Figure 2 As shown, the cloud service system can be divided into multiple logical areas according to geographical locations. The logical area can be any one of a region, an availability zone (AZ), a data center (DC) and a chassis.
[0106] For any logical area, multiple physical servers are deployed in the logical area, where the physical servers can provide cloud services to users in a variety of ways: for example, multiple virtual instances can be deployed on the physical server, such as virtual machines (VMs), containers (docker), etc. These virtual instances can be allocated to users to provide cloud services to users. For another example, the physical server can also directly provide cloud services to users in the form of a bare metal server (i.e., the physical server itself).
[0107] Furthermore, the logical area may also be deployed with multiple Top of Rack (TOR) switches and multiple aggregation switches. Then, in the logical area, the server can be connected to the cache device outside the logical area through the TOR switch and the aggregation switch in sequence, which is equivalent to the server in the logical area being indirectly connected to the cache device outside the logical area, and the two can communicate. For example, Figure 2 In the cloud service system shown, cache device C0, cache device C1, cache device C2 and cache device C3 are deployed. Logical area 0 is deployed with server E00, server E01, server E10, server E11, TOR switch T0, TOR switch T1, aggregation switch A0 and aggregation switch A1. Among them, one end of switch T0 is connected to server E00 and server E01, the other end of switch T0 is connected to one end of aggregation switch A0, and the other end of aggregation switch A0 is connected to cache device C0 and cache device C1. One end of switch T1 is connected to server E10 and server E11, the other end of switch T1 is connected to one end of aggregation switch A1, and the other end of aggregation switch A1 is connected to cache device C2 and cache device C3. In this way, server E00 can communicate with cache device C0 through TOR switch T0 and aggregation switch A0 in sequence, and the same is true for other servers, which will not be repeated here.
[0108] In the cloud service system, multiple cache devices are deployed outside all logical areas. The cache devices can be physical servers or switches (for example, cache devices can also be called core switches). These cache devices can serve as communication bridges between servers in different logical areas and can be used to cache data transmitted across logical areas. Still as in the above example, when server E00 in logical area 0 needs to send data to server E20 in logical area 1, server E00 can achieve data transmission through the following path: server E00 first sends the data to cache device C1 for temporary storage, and cache device C1 then sends the data to server E20 for processing. Of course, server E00 can also achieve data transmission through other paths.
[0109] It is worth noting that an arbitrator is set in each logical area. The arbitrator can be a physical server, an offload card or a heterogeneous card, etc., wherein the offload card can be presented in the form of a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), etc., and the heterogeneous card can be presented in the form of a graphics processing unit (GPU), a network interface controller (NIC), etc. Each arbitrator can manage all servers in its logical area, for example, to implement path planning, time slot allocation and time synchronization for data transmission, etc. In order to further understand the process of arbitrator management of servers, the following is combined with Figure 3 This process is further described. Figure 3 A flow chart of a data transmission method provided in an embodiment of the present application, which can be performed as follows Figure 2 The cloud service system shown is implemented as Figure 3 As shown, the method includes:
[0110] 301. A first arbitrator receives a fourth request from a first server, where the fourth request is used to instruct sending first data to a second server, and the first arbitrator manages the first server.
[0111] In a cloud service system, when a first server in a first logical area needs to send first data to a second server in a second logical area (i.e., data transmission across logical areas needs to be realized), a fourth request may be first sent to a first arbitrator that manages all servers in the first logical area. The fourth request may include information such as the amount of the first data and an identifier of the second server, so the fourth request may be used to instruct sending the first data to the second server.
[0112] For example, when server E00 in logical zone 0 needs to send data to server E20 in logical zone 1, it may first send a request to arbitrator Z0 in logical zone 0, so that arbitrator Z0 may determine based on the request that server E00 needs to send data to server E20.
[0113] 302. The first arbitrator determines a first cache device from a plurality of cache devices according to a fourth request, and generates a first path.
[0114] The first arbitrator can centrally process multiple requests received within a certain time period, and these requests may include requests from servers in the first logical area and requests from various cache devices. Then, in the current time period (i.e., the time period in which the first arbitrator receives the fourth request), in addition to the first arbitrator receiving the fourth request from the first server, multiple situations may occur, which will be introduced below:
[0115] (1) In the current time period, if the first arbitrator receives no other requests from the first server to the first arbitrator (these requests are used to indicate that the first server needs to send data to other servers in the first logical area) except for the fourth request, the first arbitrator may directly process the fourth request.
[0116] (2) In the current time period, if the first arbitrator receives the seventh request sent by the first server to the first arbitrator in addition to the fourth request, the seventh request may include information such as the amount of the second data and the identifier of the third server, so the seventh request may be used to indicate that the first server needs to send the second data to the third server in the first logical area. After parsing the fourth request and the seventh request, the first arbitrator can determine based on the information in the two requests that the first server needs to send not only the first data to the second server, but also the second data to the third server, that is, the first server needs to realize both cross-logical area data transmission and internal logical area data transmission. Then, the first arbitrator can select one of the fourth request and the seventh request for priority processing based on the preset request priority.
[0117] Generally, the priority of the seventh request is usually higher than that of the fourth request, so the first arbitrator can process the seventh request first, that is, first realize the data transmission between the first server and the third server, and then process the fourth request. Still as in the above example, in the current time period, arbitrator Z0 receives two requests from server E00, request 1 is used to indicate sending data to server E20, and request 2 is used to indicate sending data to server E10, so arbitrator Z0 will first process request 2, that is, arbitrator Z0 can plan the following path based on request 2: server E00→TOR switch T0→aggregation switch A0→TOR switch T1→server E10, and notify server E00 of the path, so that server E00 sends data to server E10 along the path. Thereafter, arbitrator Z0 processes request 1 again.
[0118] (3) After completing (1) or (2), the first arbitrator may process the fourth request. Specifically, based on the amount of the first data obtained by parsing the fourth request and information such as the identifier of the second server, the first arbitrator may determine that the first server needs to send the first data to the second server. Then, the first arbitrator may determine a first cache device from multiple cache devices for temporarily storing the first data. Specifically, there may be multiple situations in which the first arbitrator determines the first cache device, which will be introduced below respectively:
[0119] (3.1) In the current time period, if the first arbitrator receives no request from any cache device except the fourth request, the first arbitrator can determine that each cache device can send data to all logical areas (i.e., it considers that all cache devices are available), so the first arbitrator can randomly select a cache device from all cache devices as the first cache device. Still as in the above example, suppose that in the current time period, arbitrator Z0 only receives the request from server E00 and does not receive any request from any cache device. Arbitrator Z0 then considers that cache devices C0, C1, C2 and C3 are all available, and selects one cache device from them to temporarily store the data of server E00.
[0120] (3.2) In the current time period, in addition to receiving the fourth request, if the first arbitrator also receives the fifth request sent by the second cache device to the first arbitrator, and the fifth request is used to indicate that data is to be sent to a server in the first logical area, then the first arbitrator can determine based on the request that each cache device can send data to all logical areas (i.e., it is considered that all cache devices are available), so the first arbitrator can randomly select a cache device from all cache devices as the first cache device. Still as in the above example, suppose that in the current time period, arbitrator Z0 receives a request from server E00 and also receives a request sent by cache device C0, and the request is used to indicate that cache device C0 is ready to send data to server E10, so arbitrator Z0 considers that cache device C0, cache device C1, cache device C2 and cache device C3 are all available, and selects a cache device from them to temporarily store the data of server E00.
[0121] (3.3) In the current time period, in addition to receiving the fourth request, if the first arbitrator also receives the fifth request sent by the second cache device to the first arbitrator, and the fifth request is used to instruct the second cache device to refuse to send data to multiple logical areas (for example, all logical areas of the cloud service system), the first arbitrator can determine that the second cache device is unavailable based on the request, and can remove the second cache device from all cache devices, and randomly select a cache device from the remaining cache devices as the first cache device. Still as in the above example, suppose that in the current time period, arbitrator Z0 receives a request from server E00 and also receives a request sent by cache device C0, and the request is used to instruct cache device C0 to refuse to send data to all logical areas, so arbitrator Z0 believes that cache device C0 is unavailable for all logical areas, while cache devices C1, cache device C2 and cache device C3 are all available, and selects a cache device from these available cache devices to temporarily store the data of server E00.
[0122] (3.4) In the current time period, in addition to receiving the fourth request, if the first arbitrator also receives the sixth request sent by the second cache device to the first arbitrator, and the sixth request is used to instruct the second cache device to refuse to send data to the second logical area, the first arbitrator can determine that the second cache device is unavailable for the second logical area based on the request, and can remove the second cache device from all cache devices, and randomly select a cache device from the remaining cache devices as the first cache device. Still as in the above example, suppose that in the current time period, arbitrator Z0 receives a request from server E00 and a request sent by cache device C0, and the request is used to instruct cache device C0 to refuse to send data to logical area 1, so arbitrator Z0 believes that cache device C0 is unavailable for logical area 1, while cache devices C1, cache devices C2 and cache devices C3 are all available, and selects a cache device from these available cache devices to temporarily store the data of server E00.
[0123] After determining the first cache device, the first arbitrator can generate a first path for transmitting the first data, the first path starting from the first server and ending at the first cache device. Still as in the above example, arbitrator Z0 selects cache device C1 to temporarily store the data of server E00, so arbitrator Z0 can plan the following path: server E00→TOR switch T0→aggregation switch A0→cache device C1.
[0124] It should be understood that in the aforementioned (2), only two requests received by the first arbitrator from the first server are used for schematic illustration. In actual applications, the first arbitrator may receive more requests from the first server, for example, three requests, four requests, etc.
[0125] It should also be understood that in the above (3.2), (3.3) and (3.4), only the first arbitrator receives the request of the second cache device for schematic illustration. In actual applications, the first arbitrator may receive requests from more cache devices. For example, the first arbitrator receives not only the request of the second cache device, but also the request of the third cache device and the request of the fourth cache device, and so on.
[0126] It should also be understood that the foregoing is only a schematic illustration of the first arbitrator randomly selecting a cache device for temporarily storing data of the server E00. In actual applications, the first arbitrator can also make a selection based on a preset rule, which may be the available cache size of the cache device, the number size of the cache device, etc.
[0127] 303. The first arbitrator sends a fourth response to the first server, where the fourth response is used to instruct the first server to send the first data to be sent to the second server to the first cache device along the first path.
[0128] After obtaining the first path, the first arbitrator sends a fourth response to the first server. The fourth application may include information such as the first path and the identifier of the first cache device. Therefore, the fourth response can be used to instruct the first server to send the first data to be sent to the second server to the first cache device along the first path.
[0129] 304. The first cache device receives, on a first path, first data from a first server to be sent to a second server.
[0130] After receiving the fourth response, the first server can parse the fourth response to obtain information such as the first path and the identifier of the first cache device. Then, based on this information and in accordance with the instructions of the first arbitrator, the first server can send the first data to the first cache device along the first path. It is worth noting that the first server can make the first cache device determine that the first data is data to be sent to the second server (i.e., data to be sent to the second server) in some way. For example, the first server can make the first data carry the identifier of the second server.
[0131] Then, after receiving the first data from the first server on the first path, the first cache device can obtain the identifier of the second server by parsing the first data. Then, the first cache device can determine the second arbitrator that manages the second server based on the identifier of the second server. Still as in the above example, after the cache device C1 receives the data from the server E00, since the data carries the identifier of the server E20, the cache device C1 can determine the logical area 1 where the server E20 is located and the arbitrator Z1 in the logical area 1 based on the identifier.
[0132] Furthermore, after receiving the first data from the first server on the first path, the first cache device can also determine the amount of data to be sent to the multiple logical areas and the amount of data to be sent to the second logical area based on the first data. As in the above example, after receiving the data from the server E00, the cache device C1 can calculate the total amount of data that it needs to send to logical area 0, logical area 1, logical area 2, and logical area 3 (assuming Figure 2 The cloud service system shown includes 4 logical areas), and the amount of data that needs to be sent to logical area 1.
[0133] After obtaining the amount of data to be sent to the multiple logical areas and the amount of data to be sent to the second logical area, the first cache device can compare the amount of data to be sent to the multiple logical areas with the preset data amount threshold, and determine whether the amount of data to be sent to the second logical area is within the preset value range. The preset data amount threshold can be determined by the following formula:
[0134] T fc =α(B-kT flight -B shared_used )
[0135] In the formula, α is a preset value, B is the total amount of data that can be cached by the first cache device, k is the network bandwidth, and B shared_used is the amount of data in all shared caches used by the first cache device, T flight The clearance time is the sum of the time it takes for the first cache device to communicate with the first arbitrator, the time it takes for the first arbitrator to process information, the time it takes for the first arbitrator to communicate with the first server, and the time it takes for the first server to communicate with the first cache device. Figure 4 For further introduction. Figure 4 A schematic diagram of the headroom time provided in the embodiment of the present application, such as Figure 4 As shown, the clearance time of the cache device C1 includes the communication time t1 between the cache device C1 and the arbitrator Z0, the time t2 for the arbitrator Z0 to process information (e.g., request, etc.), the communication time t3 between the arbitrator Z0 and the server E00, and the communication time t4 between the server E00 and the cache device C1.
[0136] The right endpoint of the preset value range is usually a preset value (which can be set according to actual needs), and the left endpoint is determined according to the time when the first cache device communicates with the first arbitrator, the time when the first arbitrator processes information, the time when the first arbitrator communicates with the first server, and the time when the first server communicates with the first cache device. For example, the left endpoint can be T flight .
[0137] Then, in the first cache device, if the amount of data to be sent to the multiple logical areas is less than or equal to the preset data amount threshold, it means that the cache of the first cache device has not overflowed. If the amount of data to be sent to the second logical area is within the preset value range, it means that the amount of data to be sent from the first cache device to the second logical area is neither too much nor too little, and is within the normal processable range, that is, the port queue of the second logical area is normal.
[0138] In summary, after the first cache device compares the amount of data to be sent to the multiple logical areas with the preset data amount threshold and determines whether the amount of data to be sent to the second logical area is within the preset value range, the following three situations may occur:
[0139] (1) If the amount of data to be sent to the multiple logical areas is less than or equal to a preset data amount threshold, and the amount of data to be sent to the second logical area is within a preset value range, the first cache device can determine that the cache of the first cache device has not overflowed, and the port length of the second logical area is normal, so the first cache device sends a first request to the second arbitrator, that is, executes step 305, which is not expanded here.
[0140] (2) If the amount of data to be sent to multiple logical areas is greater than the preset data amount threshold, the first cache device may determine that there is a risk of overflow in the cache of the first cache device, so the first cache device sends a second request to the arbitrators in the multiple logical areas, and the second request is used to instruct the first cache device to refuse to send data to the multiple logical areas. Then, after the arbitrators of the multiple logical areas receive the second request, each arbitrator will prohibit the server it manages from sending data to each logical area through the first cache device (considering that the first cache device is unavailable to all logical areas). For example, after the first arbitrator receives the second request from the first cache device (the second request is used to instruct the first cache device to refuse to send data to multiple logical areas), at this time, for the first arbitrator, it is usually the next time period. When the first arbitrator centrally processes multiple requests in the next time period, it can remove the first cache device from the multiple cache devices according to the second request. The process can refer to the relevant description part of the aforementioned (3.3), which will not be repeated here. It should be noted that the other arbitrators will also perform the same process as the first arbitrator, which will not be repeated here.
[0141] Still as in the above example, after cache device C1 calculates the total amount of data that it needs to send to logical area 0, logical area 1, logical area 2 and logical area 3, if it determines that the total amount of data is greater than the preset data amount threshold, it sends a request to arbitrator Z0, arbitrator Z1, arbitrator Z2, and arbitrator Z3 to let these arbitrators know that cache device C1 refuses to send data to all logical areas. Therefore, arbitrator Z0, arbitrator Z1, arbitrator Z2, and arbitrator Z3 will all consider cache device C1 to be unavailable. If there is a need for data transmission across logical areas in the next time period, each arbitrator will no longer consider cache device C1 when selecting a cache device.
[0142] (3) If the amount of data to be sent to the multiple logical areas is less than or equal to the preset data amount threshold, and the amount of data to be sent to the second logical area is outside the preset value range, the first cache device can determine that the port captain of the second logical area is abnormal, and then send a third request to the arbitrators in the multiple logical areas, and the third request is used to instruct the first cache device to refuse to send data to the second logical area. Then, after the arbitrators of the multiple logical areas receive the third request, each arbitrator will prohibit the server it manages from sending data to the second logical area through the first cache device (considering that the first cache device is unavailable to the second logical area). For example, after the first arbitrator receives the third request from the first cache device (the second request is used to instruct the first cache device to refuse to send data to the second logical area), at this time, for the first arbitrator, it is usually the next time period. When the first arbitrator centrally processes multiple requests in the next time period, it can remove the first cache device from the multiple cache devices according to the third request. The process can refer to the relevant description part of the aforementioned (3.3), which will not be repeated here. It should be noted that the other arbitrators will also perform the same process as the first arbitrator, which will not be repeated here.
[0143] Still as in the above example, after cache device C1 calculates the amount of data it needs to send to logical area 1, if it determines that the amount of data is outside the preset value range, it sends a request to arbitrator Z0, arbitrator Z1, arbitrator Z2, and arbitrator Z3 to let these arbitrators know that cache device C1 refuses to send data to logical area 1. Therefore, arbitrator Z0, arbitrator Z1, arbitrator Z2, and arbitrator Z3 will all consider cache device C1 to be unavailable for logical area 1. If there is a need for data transmission across logical areas in the next time period (the target logical area to be sent is logical area 1), each arbitrator will no longer consider cache device C1 when selecting a cache device.
[0144] 305. The first cache device sends a first request to a second arbitrator that manages the second server, where the first request is used to instruct sending the first data to the second server.
[0145] If the amount of data to be sent to multiple logical areas is less than or equal to a preset data amount threshold, and the amount of data to be sent to the second logical area is within a preset value range, the first cache device sends a first request to the second arbitrator. The first request includes information such as the identifier of the second server and the data amount of the first data. Therefore, the first request can be used to indicate that the first cache device needs to send the first data to the second server.
[0146] 306. The second arbitrator generates a second path according to the first request.
[0147] Similarly, during the period when the second arbitrator receives the first request from the first cache device, multiple situations may occur, which will be described below respectively:
[0148] (1) During the time period, if the second arbitrator receives no requests from other servers in the second logical area other than the first request (these requests are used to indicate that data needs to be sent to the second server), the second arbitrator may directly process the first request.
[0149] (2) During this time period, if the second arbitrator receives, in addition to the first request, an eighth request sent by the fourth server in the second logical area, the eighth request may include information such as the amount of the third data and the identifier of the second server, so the eighth request may be used to indicate that the fourth server needs to send the third data to the second server. After the second arbitrator parses the first request and the eighth request, it can be determined based on the information in the two requests that not only the first cache device needs to send the first data to the second server, but also the fourth server needs to send the third data to the second server, that is, the second server needs to realize both cross-logical area data transmission and intra-logical area data transmission. Then, the second arbitrator can select one of the first request and the eighth request for priority processing based on the preset request priority.
[0150] Generally, the priority of the eighth request is usually higher than that of the first request, so the second arbitrator can process the eighth request first, that is, first realize the data transmission between the fourth server and the second server, and then process the first request. Still as in the above example, in the current time period, the arbitrator Z1 receives the request of the cache device C1 and the request of the server E30. Both requests are used to indicate sending data to the server E20, so the arbitrator Z1 will first process the request of the server E30, that is, the arbitrator Z1 can plan the following path based on the request of the server E30: server E30→TOR switch T3→aggregation switch A2→TOR switch T2→server E20, and notify the server E30 of the path, so that the server E30 sends the data to the server E20 along the path. Thereafter, the arbitrator Z1 processes the request of the cache device C1 again.
[0151] (3) After completing (1) or (2), the first arbitrator can process the first request. Specifically, based on the identifier of the second server obtained by parsing the first request and information such as the amount of the first data, the second arbitrator can determine that the first cache device needs to send the first data to the second server. Therefore, the second arbitrator can generate a second path for transmitting the first data. The second path starts at the first cache device and ends at the second server. Still like the above example, the arbitrator Z1 can plan the following path: cache device C1→aggregation switch A2→TOR switch T2→server E20.
[0152] It should be understood that in the aforementioned (2), only the second arbitrator receives the request of the first cache device and the request of the fourth server (both directed to the second server) for schematic illustration. In actual application, the first arbitrator may receive more requests directed to the second server.
[0153] 307. The first cache device receives a first response from the second arbitrator, where the first response is used to indicate the second path.
[0154] After obtaining the first path, the second arbitrator sends a first response to the first cache device. The first response may include information such as the second path and the identifier of the second server. Therefore, the first response can be used to instruct the first cache device to send the first data to the second server along the second path.
[0155] 308. The first cache device sends the first data to the second server along the second path indicated by the first response.
[0156] After receiving the first response, the first cache device can parse the first response to obtain information such as the second path and the identifier of the second server. Then, based on this information and according to the instruction of the second arbitrator, the first cache device can send the first data to the second server along the second path. At this point, the data transmission between the first server and the second server across logical areas is completed.
[0157] It is worth noting that the first arbitrator and all servers in the first logical area need to maintain time synchronization (for example, the first arbitrator can adjust the local time of each server in the first logical area so that the time of all servers and the first arbitrator are synchronized), and the second arbitrator and all servers in the second logical area also need to maintain time synchronization, and the same applies to the remaining logical areas. Furthermore, time synchronization does not need to be maintained between different logical areas. For example, the local time of the devices in the first logical area and the local time of the devices in the second logical area may not be synchronized, and so on. Then, for any cache device, since the cache device is connected to each logical area, the time of the port of the cache device connected to any logical area needs to be synchronized with the local time of the devices in the logical area, and the local time of the devices in the logical area must be maintained. For example, if Figure 5 As shown ( Figure 5 Another structural diagram of the cloud service system provided in an embodiment of the present application), the first port of the first cache device is connected to the first logical area, and the time of the first port needs to be synchronized with the time of all devices in the first logical area, the second port of the first cache device is connected to the second logical area, and the time of the second port needs to be synchronized with the time of all devices in the second logical area, and so on, and the same applies to the remaining cache devices.
[0158] In an embodiment of the present application, when the first server needs to send the first data to the second server, the first arbitrator that manages the first server can plan a first path for the first server so that the first server sends the first data to the first cache device along the first path for caching. Then, the first cache device can send a first request to the second arbitrator that manages the second server to notify the first cache device that the first data needs to be sent to the second server, so the second arbitrator can plan a second path for the first cache device so that the first cache device sends the first data to the second server along the second path. In the above process, a first cache device is provided between the first server and the second server as a forwarding point for temporarily storing data. Then, the data transmission path between the first server and the second server can be divided into two parts. The first path between the first server and the first cache device can be planned by the first arbitrator that manages the first server, and the second path between the first cache device and the second server can be planned by the second arbitrator that manages the second server. It can be seen that a first arbitrator is provided at the first server as the data sending end, and a second arbitrator is provided at the second server as the data receiving end. The two arbitrators can jointly complete the path scheduling between the first server and the second server and share the scheduling pressure. Therefore, there is no need to make excessively high requirements on the computing power of the arbitrator, and the problem of network congestion can be effectively solved.
[0159] Furthermore, the embodiments of the present application do not require any hardware improvements, but only require improvements to the software of the devices (arbitrators, cache devices, etc.). The cost required is relatively low, which is conducive to the large-scale expansion of the entire cloud service system and can promote the commercial implementation of the cloud service system.
[0160] Furthermore, in the embodiment of the present application, the arbitrator and the server in the same logical area can perform time synchronization, while the devices in multiple logical areas do not need to perform time synchronization. The cache device connects the ports of different logical areas to the devices in different logical areas to achieve time synchronization and maintain the clock of each logical area. It can be seen that the entire cloud service system does not need to perform global time synchronization, and only needs to achieve local time synchronization.
[0161] Furthermore, when storing and forwarding data, the cache device can determine whether its own cache is overflowing and whether the port captain of each logical area is normal based on the amount of data to be sent to each logical area, thereby effectively ensuring its own normal operation and improving the stability of the entire cloud service system.
[0162] The above is a specific description of the data transmission method provided in the embodiment of the present application. The following will introduce the cache device and arbitrator provided in the embodiment of the present application. Figure 6 A schematic diagram of a cache device provided in an embodiment of the present application is shown in FIG. Figure 6 As shown, the cache device serves as a first cache device, and the first cache device includes:
[0163] A receiving module 601 is used to receive first data to be sent to a second server from a first server on a first path, where the first path is determined by a first arbitrator that manages the first server;
[0164] A sending module 602, configured to send a first request to a second arbitrator that manages a second server, wherein the first request is used to instruct sending first data to the second server;
[0165] A receiving module 601 is used to receive a first response from a second arbitrator, where the first response is used to indicate a second path and is obtained by the second arbitrator processing the first request;
[0166] The sending module 602 is used to send the first data to the second server along the second path indicated by the first response.
[0167] In a possible implementation manner, the first server and the first arbitrator are located in a first logical area, and the second server and the second arbitrator are located in a second logical area.
[0168] In a possible implementation, the first logical area and the second logical area are regions, availability zones, data centers, or chassis.
[0169] In a possible implementation, the first cache device further includes: a processing module 603, configured to determine, based on the first data, an amount of data to be sent to the plurality of logical areas and an amount of data to be sent to the second logical area, the plurality of logical areas including the first logical area and the second logical area;
[0170] The sending module 602 is configured to send a first request to the second arbitrator if the amount of data to be sent to the multiple logical areas is less than or equal to a preset data amount threshold and the amount of data to be sent to the second logical area is within a preset value range.
[0171] In a possible implementation, the sending module 602 is further used to send a second request to the arbitrators in the multiple logical areas if the amount of data to be sent to the multiple logical areas is greater than a data amount threshold, and the second request is used to instruct the first cache device to refuse to send data to the multiple logical areas.
[0172] In one possible implementation, the sending module 602 is further used to send a third request to the arbitrator in the multiple logical areas if the amount of data to be sent to the multiple logical areas is less than or equal to the data amount threshold, and the amount of data to be sent to the second logical area is outside the value range, and the third request is used to instruct the first cache device to refuse to send data to the second logical area.
[0173] In a possible implementation, the data volume threshold is determined according to the time when the first cache device communicates with the first arbitrator, the time when the first arbitrator processes information, the time when the first arbitrator communicates with the first server, and the time when the first server communicates with the first cache device.
[0174] In a possible implementation, the left endpoint of the value range is determined according to the time when the first cache device communicates with the first arbitrator, the time when the first arbitrator processes information, the time when the first arbitrator communicates with the first server, and the time when the first server communicates with the first cache device.
[0175] In a possible implementation, the processing module 603 is configured to determine a second arbitrator that manages the second server according to the identifier of the second server; and the sending module 602 is configured to send the first request to the second arbitrator.
[0176] In a possible implementation manner, the first cache device is a switch or a server.
[0177] Figure 7 A structural diagram of an arbitrator provided in an embodiment of the present application, such as Figure 7 As shown, the arbitrator serves as a first arbitrator, and the first arbitrator includes:
[0178] The receiving module 701 is used to receive a fourth request from the first server, where the fourth request is used to instruct to send the first data to the second server, and the first arbitrator manages the first server;
[0179] The processing module 702 is used to determine a first cache device from a plurality of cache devices according to a fourth request, and generate a first path;
[0180] The sending module 703 is used to send a fourth response to the first server, where the fourth response is used to instruct the first server to send the first data to be sent to the second server to the first cache device along the first path.
[0181] In a possible implementation manner, the first server and the first arbitrator are located in a first logical area, and the second server and a second arbitrator that manages the second server are located in a second logical area.
[0182] In a possible implementation, the first logical area and the second logical area are regions, availability zones, data centers, or chassis.
[0183] In a possible implementation, the receiving module 701 is further used to receive a fifth request from the second cache device, where the fifth request is used to instruct the second cache device to refuse to send data to multiple logical areas, where the multiple logical areas include a first logical area and a second logical area; the processing module 702 is used to remove the second cache device from the multiple cache devices according to the fourth request and the fifth request, and select the first cache device from the remaining cache devices.
[0184] In a possible implementation, the receiving module 701 is further used to receive a sixth request from the second cache device, where the sixth request is used to instruct the second cache device to refuse to send data to the second logical area; the processing module 702 is used to remove the second cache device from the multiple cache devices according to the fourth request and the sixth request, and select the first cache device from the remaining cache devices.
[0185] In a possible implementation, the receiving module 701 is further used to receive a second request from the first cache device, where the second request is used to instruct the first cache device to refuse to send data to multiple logical areas; the processing module 702 is further used to remove the first cache device from the multiple cache devices based on the second request.
[0186] In a possible implementation, the receiving module 701 is further used to receive a third request from the first cache device, where the third request is used to instruct the first cache device to refuse to send data to the second logical area; the processing module 702 is further used to remove the first cache device from the multiple cache devices according to the third request.
[0187] In a possible implementation, the receiving module 701 is further used to receive a seventh request from the first server, where the seventh request is used to indicate sending second data to the third server, and the first arbitrator and the third server are located in the first logical area; the processing module 702 is further used to select one of the fourth request and the seventh request for priority processing based on a preset request priority.
[0188] In a possible implementation manner, the first data carries an identifier of the second server.
[0189] In a possible implementation manner, the first cache device and the second cache device are switches or servers.
[0190] Figure 8 Another structural diagram of the arbitrator provided in the embodiment of the present application is as follows: Figure 8 As shown, the arbitrator serves as a second arbitrator, and the second arbitrator includes:
[0191] A receiving module 801 is used to receive a first request from a first cache device, where the first request is used to instruct to send first data to a second server;
[0192] A processing module 802 is used to generate a second path according to the first request;
[0193] The sending module 803 is used to send a first response to the first cache device, where the first response is used to instruct the first cache device to send the first data to the second server along the second path.
[0194] In a possible implementation manner, the second server and the second arbitrator are located in the second logical area.
[0195] In a possible implementation, the second logical area is a region, an availability zone, a data center, or a chassis.
[0196] In a possible implementation, the receiving module 801 is further used to receive an eighth request from a fourth server, where the eighth request is used to indicate sending third data to the second server, and the second arbitrator and the fourth server are located in the second logical area; the processing module 802 is further used to select one of the first request and the eighth request for priority processing based on a preset request priority.
[0197] In a possible implementation manner, the first cache device is a switch or a server.
[0198] It should be noted that the information interaction, execution process, etc. between the modules / units of the above-mentioned device are based on the same concept as the method embodiment of the present application, and the technical effects they bring are the same as those of the method embodiment of the present application. For specific contents, please refer to the description in the method embodiment shown above in the embodiment of the present application, and will not be repeated here.
[0199] Fig. 9 Another structural diagram of the cache device provided in the embodiment of the present application. Fig. 9 As shown, the cache device in the embodiment of the present application can be used as a first cache device. An embodiment of the first cache device may include one or more central processors 901, a memory 902, an input and output interface 903, a wired or wireless network interface 904, and a power supply 905.
[0200] The memory 902 may be a temporary storage or a permanent storage. Furthermore, the central processor 901 may be configured to communicate with the memory 902 and execute a series of instruction operations in the memory 902 on the first cache device.
[0201] In this embodiment, the CPU 901 can execute the aforementioned Figure 3 The operations performed by the first cache device in the illustrated embodiment will not be described in detail herein.
[0202] In this embodiment, the specific functional module division in the central processing unit 901 can be the same as the above Figure 6 The division method of the receiving module, processing module, sending module and other modules described in is similar and will not be repeated here.
[0203] Fig.10 Another structural diagram of the arbitrator provided in the embodiment of the present application. Fig.10 As shown, the arbitrator in the embodiment of the present application can be used as a first arbitrator. An embodiment of the first arbitrator may include one or more central processors 1001, a memory 1002, an input and output interface 1003, a wired or wireless network interface 1004, and a power supply 1005.
[0204] The memory 1002 may be a temporary storage or a permanent storage. Furthermore, the central processor 1001 may be configured to communicate with the memory 1002 and execute a series of instruction operations in the memory 1002 on the first arbitrator.
[0205] In this embodiment, the CPU 1001 can execute the aforementioned Figure 3 The operations performed by the first arbitrator in the illustrated embodiment will not be described in detail herein.
[0206] In this embodiment, the specific functional module division in the central processing unit 1001 can be the same as the above Figure 7The division method of the receiving module, processing module, sending module and other modules described in is similar and will not be repeated here.
[0207] Fig.11 Another structural diagram of the arbitrator provided in the embodiment of the present application. Fig.11 As shown, the arbitrator in the embodiment of the present application can be used as a second arbitrator. An embodiment of the second arbitrator may include one or more central processors 1101, a memory 1102, an input and output interface 1103, a wired or wireless network interface 1104, and a power supply 1105.
[0208] The memory 1102 may be a temporary storage or a permanent storage. Furthermore, the central processor 1101 may be configured to communicate with the memory 1102 and execute a series of instruction operations in the memory 1102 on the second arbitrator.
[0209] In this embodiment, the CPU 1101 may execute the aforementioned Figure 3 The operations performed by the second arbitrator in the illustrated embodiment will not be described in detail herein.
[0210] In this embodiment, the specific functional module division in the central processor 1101 can be the same as the above Figure 8 The division method of the receiving module, processing module, sending module and other modules described in is similar and will not be repeated here.
[0211] The embodiment of the present application also relates to a computer storage medium, including computer readable instructions. When the computer readable instructions are executed, the following is implemented: Figure 3 The steps performed by the first cache device in the embodiment shown in the figure may be implemented as follows: Figure 3 The steps performed by the first arbitrator in the embodiment shown, or, to implement the following Figure 3 The steps performed by the second cache device in the illustrated embodiment.
[0212] The present application also relates to a computer program product comprising instructions, which, when executed on a computer, causes the computer to execute Figure 3 The steps performed by the first cache device in the embodiment shown in the figure, or performing the steps as follows Figure 3 The steps performed by the first arbitrator in the embodiment shown, or, performing the steps as follows Figure 3 The steps performed by the second cache device in the illustrated embodiment.
[0213] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.
[0214] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.
[0215] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0216] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.
[0217] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk.
Claims
1. A data transmission method, characterized in that: The method comprises: The first cache device receives first data to be sent to the second server from the first server on a first path, wherein the first path is determined by a first arbitrator that manages the first server; The first cache device sends a first request to a second arbitrator that manages the second server, where the first request is used to instruct sending the first data to the second server; The first cache device receives a first response from the second arbitrator, where the first response is used to indicate a second path, and the first response is obtained by the second arbitrator processing the first request; The first cache device sends the first data to the second server along a second path indicated by the first response.
2. The method according to claim 1, characterized in that The first server and the first arbitrator are located in a first logical zone, and the second server and the second arbitrator are located in a second logical zone.
3. The method according to claim 2, characterized in that The first logical area and the second logical area are regions, availability zones, data centers, or chassis.
4. The method according to claim 2 or 3, characterized in that: Before the first cache device sends the first request to the second arbitrator, the method further includes: The first cache device determines, based on the first data, an amount of data to be sent to a plurality of logical areas and an amount of data to be sent to the second logical area, the plurality of logical areas including the first logical area and the second logical area; The first cache device sending the first request to the second arbitrator includes: If the amount of data to be sent to the multiple logical areas is less than or equal to a preset data amount threshold, and the amount of data to be sent to the second logical area is within a preset value range, the first cache device sends a first request to the second arbitrator.
5. The method according to claim 4, characterized in that The method further comprises: If the amount of data to be sent to the multiple logical areas is greater than the data amount threshold, the first cache device sends a second request to the arbitrators in the multiple logical areas, where the second request is used to instruct the first cache device to refuse to send data to the multiple logical areas.
6. The method according to claim 4, characterized in that The method further comprises: If the amount of data to be sent to the multiple logical areas is less than or equal to the data amount threshold, and the amount of data to be sent to the second logical area is outside the value range, the first cache device sends a third request to the arbitrator in the multiple logical areas, and the third request is used to instruct the first cache device to refuse to send data to the second logical area.
7. The method according to claim 5 or 6, characterized in that: The data volume threshold is determined according to the time it takes for the first cache device to communicate with the first arbitrator, the time it takes for the first arbitrator to process information, the time it takes for the first arbitrator to communicate with the first server, and the time it takes for the first server to communicate with the first cache device.
8. The method according to claim 5 or 6, characterized in that: The left endpoint of the value range is determined according to the time when the first cache device communicates with the first arbitrator, the time when the first arbitrator processes information, the time when the first arbitrator communicates with the first server, and the time when the first server communicates with the first cache device.
9. The method according to any one of claims 1 to 3, characterized in that: The first data carries an identifier of the second server, and the first cache device sends a first request to a second arbitrator that manages the second server, including: The first cache device determines a second arbitrator that manages the second server according to the identifier of the second server; The first cache device sends a first request to the second arbitrator.
10. The method according to any one of claims 1 to 3, characterized in that: The first cache device is a switch or a server.
11. A data transmission method, characterized in that: The method comprises: The first arbitrator receives a fourth request from the first server, the fourth request being used to instruct sending first data to the second server, the first arbitrator managing the first server; The first arbitrator determines a first cache device from a plurality of cache devices according to the fourth request, and generates a first path; The first arbitrator sends a fourth response to the first server, where the fourth response is used to instruct the first server to send first data to be sent to the second server to the first cache device along the first path; The second arbitrator receives a first request from the first cache device, where the first request is used to instruct to send first data to the second server; The second arbitrator generates a second path according to the first request; The second arbitrator sends a first response to the first cache device, where the first response is used to instruct the first cache device to send the first data to the second server along a second path.
12. The method according to claim 11, characterized in that The first server and the first arbitrator are located in a first logical area, and the second server and a second arbitrator that manages the second server are located in a second logical area.
13. The method according to claim 12, characterized in that The first logical area and the second logical area are regions, availability zones, data centers, or chassis.
14. The method according to claim 12 or 13, characterized in that Before the first arbitrator determines the first cache device from the plurality of cache devices according to the fourth request, the method further includes: The first arbitrator receives a fifth request from a second cache device, the fifth request being used to instruct the second cache device to refuse to send data to a plurality of logical regions, the plurality of logical regions including the first logical region and the second logical region; The first arbitrator determines the first cache device from the plurality of cache devices according to the fourth request, comprising: The first arbitrator removes the second cache device from the multiple cache devices according to the fourth request and the fifth request, and selects the first cache device from the remaining cache devices.
15. The method according to claim 12 or 13, characterized in that Before the first arbitrator determines the first cache device from the plurality of cache devices according to the fourth request, the method further includes: The first arbitrator receives a sixth request from the second cache device, where the sixth request is used to instruct the second cache device to refuse to send data to the second logical area; The first arbitrator determines the first cache device from the plurality of cache devices according to the fourth request, comprising: The first arbitrator removes the second cache device from the multiple cache devices according to the fourth request and the sixth request, and selects the first cache device from the remaining cache devices.
16. The method according to any one of claims 11 to 13, characterized in that: After the first arbitrator sends a fourth response to the first server, the method further includes: The first arbitrator receives a second request from the first cache device, where the second request is used to instruct the first cache device to refuse to send data to the multiple logical areas; The first arbitrator removes the first cache device from the multiple cache devices according to the second request.
17. The method according to any one of claims 11 to 13, characterized in that: After the first arbitrator sends a fourth response to the first server, the method further includes: The first arbitrator receives a third request from the first cache device, where the third request is used to instruct the first cache device to refuse to send data to the second logical area; The first arbitrator removes the first cache device from the multiple cache devices according to the third request.
18. The method according to claim 12, characterized in that Before the first arbitrator determines the first cache device from the plurality of cache devices according to the fourth request and generates the first path, the method further includes: The first arbitrator receives a seventh request from the first server, the seventh request is used to instruct to send second data to a third server, the first arbitrator and the third server are located in the first logical area; The first arbitrator selects one of the fourth request and the seventh request for priority processing according to a preset request priority.
19. The method according to any one of claims 11 to 13, characterized in that: The first cache device and the second cache device are switches or servers.
20. The method according to any one of claims 11 to 13, characterized in that: Before the second arbitrator generates the second path according to the first request, the method further includes: The second arbitrator receives an eighth request from a fourth server, wherein the eighth request is used to instruct to send third data to the second server, and the second arbitrator and the fourth server are located in the second logical area; The second arbitrator selects one of the first request and the eighth request for priority processing according to a preset request priority.
21. A cache device, characterized in that: The cache device is used as a first cache device, and the first cache device includes: A receiving module, configured to receive first data to be sent from a first server to a second server on a first path, wherein the first path is determined by a first arbitrator that manages the first server; a sending module, configured to send a first request to a second arbitrator that manages the second server, wherein the first request is used to instruct sending the first data to the second server; The receiving module is used to receive a first response from the second arbitrator, where the first response is used to indicate a second path, and the first response is obtained by the second arbitrator processing the first request; The sending module is used to send the first data to the second server along the second path indicated by the first response.
22. A cache device, characterized in that: The cache device serves as a first cache device, and the first cache device includes a memory and a processor; the memory stores codes, and the processor is configured to execute the codes. When the codes are executed, the first cache device executes the method according to any one of claims 1 to 10.
23. An arbitrator, characterized in that: The arbitrator serves as a first arbitrator, and the first arbitrator includes a memory and a processor; the memory stores codes, and the processor is configured to execute the codes. When the codes are executed, the first arbitrator executes the method according to any one of claims 11 to 20.
24. An arbitrator, characterized in that: The arbitrator serves as a second arbitrator, and the second arbitrator includes a memory and a processor; the memory stores codes, and the processor is configured to execute the codes. When the codes are executed, the second arbitrator executes the method according to any one of claims 11 to 20.
25. A cloud service system, characterized in that: The system comprises: a first server, a second server, the first cache device according to claim 22, the first arbitrator according to claim 23, and the second arbitrator according to claim 24; The first end of the first server is connected to the first arbitrator, and the second end of the first server is connected to the first end of the first cache device; A first end of the second server is connected to the second arbitrator, and a second end of the second server is connected to a second end of the first cache device.
26. A computer storage medium, characterized in that The computer storage medium stores one or more instructions, which, when executed by one or more computers, enable the one or more computers to implement the method of any one of claims 1 to 20.
27. A computer program product, characterized in that The computer program product stores instructions, which, when executed by a computer, cause the computer to implement the method according to any one of claims 1 to 20.
Citation Information
Patent Citations
Link establishment method and equipment
CN102420764A
Data transmission method of content-centric datacenter network
CN103179037A
