Cloud Computing Elastic Scaling Design for Image Data Processing System Based on Plan-driven
Image data processing system resources are built on a private cloud.According to the characteristics of non-periodic changes of system task load and strong correlation of the task plans,and using the advantage of image data processing plans to predict business load,a user-defined index for task quantity prediction based on plans is designed,and an elastic scaling strategy and system design based on plan-driven is proposed.Through experimental environment testing,the dynamic adjustment of computing resources based on business load has been realized,and the resource utilization has been improved.At the same time,a scalable guaranteed strategy based on CPU load has been added to improve the system reliability.
cloud computingimage data processingflexible expansiondynamic load adjustment
NETL
NSTL
万方数据
