Service operation method and device based on accelerator card, electronic equipment and storage medium

Abstract

The embodiment of the invention discloses a service operation method and device based on an accelerator card, electronic equipment and a storage medium, relates to the technical field of computers, and can effectively improve the utilization rate of the accelerator card while ensuring the service response speed of the accelerator card. At least one service is deployed on the accelerator card, andthe method comprises the following steps: for each service in the at least one service, in response to an event that the service enters a frozen state, changing a process corresponding to the servicefrom an onboard memory of the accelerator card to a memory of a central processing unit (CPU); and for each service in the at least one service, in response to an event that the service enters an activated state, changing a process corresponding to the service into an onboard memory of the accelerator card from a memory of the CPU. The invention can be used for service operation of the acceleratorcard.

Description

technical field [0001] The present invention relates to the field of computer technology, in particular to a service running method, device, electronic equipment and storage medium based on an accelerator card. Background technique [0002] With the development of big data and artificial intelligence technology, many services are based on a large number of complex data calculations, but the computing power of the CPU is increasingly difficult to meet the computing requirements. Therefore, in many scenarios, the accelerator card will be used for computing acceleration. Common accelerator cards can include, for example, GPU (Graphics Processing Unit, graphics processing unit), TPU (tensor processing unit, tensor processor), NPU (Neural-network Processing Unit, embedded neural network processor), FPGA (Field Programmable Gate Array, Field Programmable Gate Array), etc. [0003] The accelerator card has powerful computing functions, but it takes a long time to initialize the p...

AI Technical Summary

Problems solved by technology

However, the number of daily requests for this service is very low, perhaps only dozens of requests a day, resulting in a large number of accelerator card resources being idle

[0005] For the above-mentioned how to effectively improve the utilization rate of the accelerator card while ensuring the response speed of the accelerator card service, there is no effective solution in related fields

Images