A task-type-aware flow queue adaptive management method in a virtualized environment

Abstract

The invention discloses a task type-aware flow queue adaptive management method in a virtualized environment. The method includes: (1) receiving an instruction from an administrator, and if it is a flow scheduling policy removal instruction, start flow scheduling removal function, if the command is a flow scheduling policy deployment command, then start the flow scheduling policy deployment function; (2) Based on the metadata information perception task type of different flows, determine whether the task type to which the flow belongs is bandwidth-sensitive or delay-sensitive ;(3) Map the streams of different types of tasks to different Qdisc queues in the kernel module at the virtualization layer to realize the isolation and management of the network transmission performance of the streams of different types of tasks; (4) Real-time statistics of the streams of different task types According to the load situation, the transmission weight between different queues is adaptively adjusted and optimized according to the load. The method can perceive the task type and adaptively manage the flow queue, effectively improving the network transmission performance of the task in a virtualized environment.

Description

technical field [0001] The invention belongs to the field of network transmission optimization in a virtualized environment, and in particular relates to a task type-aware flow queue self-adaptive management method in a virtualized environment. Background technique [0002] With the continuous maturity of cloud computing and virtualization technology and the continuous enrichment of cloud services, more and more applications are deployed in the cloud data center virtualization environment, involving various fields from finance, medical care, e-commerce, education, etc. industry. These applications are usually deployed in a distributed architecture, such as figure 1 shown. By splitting the application into sub-modules and packaging them in the form of virtual machines, each virtual machine is finally placed on different physical servers in the data center. Due to the distributed nature, the performance of data center applications is closely related to network transmission ...

AI Technical Summary

Problems solved by technology

Although the above research work has made some contributions to improving the network transmission performance and ensuring the network transmission performance requirements of different types of tasks, it has not considered the reasonable allocation of network resources between different types of network transmission tasks according to the actual load situation. Allocation, resulting in lower network transmission performance

Images