HPC cloud computing system based on cloud computing mobile network resource management

Abstract

The invention discloses an HPC cloud computing system based on cloud computing mobile network resource management, which comprises a cloud computing center, the cloud computing center comprises a virtual connector, a mobile data interaction system and a resource manager, and the cloud computing center is connected with a terminal cloud storage module; the cloud storage module stores network resource data by using a cloud server; the virtual connector comprises a task issuing module and a virtual connection module, and the task issuing module is used for issuing tasks received by the mobile data interaction system. According to the HPC cloud computing system based on cloud computing mobile network resource management, the mobile data interaction system connects the cloud computing center with the HPC through the virtual connection module of the virtual connector, the cloud computing center computes parameters of each node, and finally the HPC performs comprehensive computing, so that the computing difficulty is reduced, the computing speed is increased, and the computing efficiency is improved. The optimization of the time for executing the cloud computing task is ensured, and the reliability of the cloud platform is ensured.

Description

technical field [0001] The invention relates to the technical field of HPC cloud computing systems, in particular to an HPC cloud computing system based on cloud computing mobile network resource management. Background technique [0002] Existing network resource scheduling strategies are based on the monitoring system finding that a fixed virtual machine has insufficient or excessive resources for scheduling, without considering the overall resource requirements of the resource pool, and the adjustment ideas are relatively limited. [0003] At the same time, resource fault-tolerant scheduling assumes that a single machine fails. However, the resources used by large-scale cloud computing systems are highly dynamic and heterogeneous, and the inherent unreliable state of the resource environment makes cloud computing systems likely to fail large-scale resources. Sex is greatly increased. Therefore, we improve this and propose an HPC cloud computing system based on cloud compu...

AI Technical Summary

Problems solved by technology

[0002] Existing network resource scheduling strategies are based on the fact that the monitoring system finds that a certain fixed virtual machine has insufficient or excess resources and then schedules them. It does not consider the overall resource demand of the resource pool, and the adjustment ideas are relatively limited.

[0003] At the same time, resource fault-tolerant scheduling assumes that a single machine fails, but the resources used by large-scale cloud computing systems are highly dynamic and heterogeneous, and the inherent unreliability of the resource environment makes it possible for large-scale resource failures to occur in cloud computing systems sex greatly increased

Images