Method and system for data communication between cluster system devices

Abstract

The invention discloses a method and system for data communication between cluster system devices. The number of GPUs in each node is obtained, all the GPUs in each node are numbered, the calculation results of the GPUs are partitioned, all result data blocks are numbered according to the GPUs, each GPU sends the corresponding result data blocks to other GPUs according to the serial number of the data blocks, each GPU renews data in the nodes through a first preset rule according to the corresponding result data blocks at the moment and the result data blocks sent by other GPUs, the target GPU in each node conducts agreement stipulation operation on the GPUs, corresponding to the target GPU, in other nodes, data between the nodes are renewed according to a second preset rule, each GPU distributes the corresponding renewed result data blocks to other GPUs in the corresponding node, and each GPU processes one data block. The system network bandwidth is fully used, the loads of the GPUs are balanced, and the data communication between the GPUs can be efficiently achieved.

Description

technical field [0001] The invention relates to the field of high-performance computing, in particular to a data communication method and system between cluster system devices. Background technique [0002] In the current social data explosion, there are more and more information data, and people have higher and higher requirements for information data processing capabilities. Not only oil exploration, weather forecast, aerospace defense, scientific research, etc. require high-performance computing, but also financial and government informatization The demand for high-performance computing in a wider range of fields such as education, enterprise, online games, and artificial intelligence is growing rapidly. [0003] Computing speed is particularly important for high-performance computing. High-performance computing is developing toward multi-core and many-core, and heterogeneous parallelism is used to improve application computing speed. At present, CPU+GPU is a very mature ...

AI Technical Summary

Problems solved by technology

In a large-scale cluster system, data parallel computing is the main application method of parallel computing. However, due to the fact that computing devices for data parallel computing need to communicate with each other, and are limited by network bandwidth, bus bandwidth, etc., the communication traffic is relatively large. Large, the calculation needs a long time to wait for the transmission of data, so that the overall performance of the system cannot be improved

Therefore, the current limited transmission bandwidth can no longer meet the communication needs of system equipment, and the existing hardware communication capabilities cannot be greatly improved in a short period of time, so it is necessary to design a method to improve the communication capabilities based on the existing limited communication capabilities. A method for communication efficiency of system devices, thereby avoiding wasting computing resources by waiting for data transmission

Images