Dispatching method of virtual processor based on NUMA high-performance network cache resource affinity

Abstract

The invention discloses a dispatching method of a virtual processor based on NUMA high-performance network cache resource affinity. The dispatching method comprises the steps that under an NUMA framework, when a virtual machine network card is started, distribution of a network card cache on each NUMA node is obtained; based on the affinity relationship among the NUMA nodes, the affinity of each NUMA node towards the network card cache is obtained; a target NUMA node is determined by combining the distribution of the network card cache on each NUMA node and the affinity of each NUMA node towards the network card cache; the virtual processor is dispatched to a CPU of the target NUMA node. The dispatching method of the virtual processor based on NUMA high-performance network cache resource affinity resolves the problem that under the NUMA framework, the affinity of a VCPU of a virtual machine and the network card cache is not optimal, so that the speed of processing network data packages of the virtual machine network card is low.

Description

technical field [0001] The invention relates to the field of computer system virtualization, in particular to a virtual processor scheduling method based on NUMA high-performance network cache resource affinity. Background technique [0002] Virtualization technology usually integrates computing or storage functions that originally required multiple physical devices into a relatively powerful physical server, thereby realizing the integration and redistribution of hardware resources and improving the utilization of hardware devices. Cloud computing and data center construction play a very important role. [0003] A virtual machine monitor refers to a software management layer that exists between the hardware and the traditional operating system. Its main function is to manage real physical devices, such as physical CPUs, memory, etc., and abstract the underlying hardware into corresponding virtual device interfaces. , so that multiple operating systems can get the virtual h...

AI Technical Summary

Problems solved by technology

[0010] The existing Xen kernel supports several configuration methods for NUMA, including fully localized memory (NUMA aware placement) and memory configuration to several nodes, and then the virtual processor (VCPU) is scheduled to the assigned node (NUMA aware scheduling ), but this also does not fully analyze the memory resources of the virtual machine from the perspective of memory distribution, so that the affinity between the VCPU and the memory is still not optimal, which will inevitably affect the speed at which the VCPU processes memory

[0011] Therefore, those skilled in the art are committed to developing a virtual processor scheduling method based on NUMA high-performance network cache resource affinity to solve the problem that the rate of VCPU processing network data packets is not optimal

Images