Virtual machine monitoring method and apparatus

Abstract

The invention provides a virtual machine monitoring method, comprising: starting and running an operating system of a passenger plane through a kernel of a virtual machine; intercepting all system calls of the operating system of the passenger plane in the kernel of the virtual machine, and obtaining process information related with the system calls; according to the process information, judging whether processes initiating the system calls are positioned in a preset monitoring list, and if so, at least recording entry time and exit time of the system calls; and according to the entry time and the exit time, making statistics on time distribution for performing the system calls by related processes and outputting the time distribution in a visual form. In addition, the invention further provides a virtual machine monitoring apparatus. The virtual machine monitoring method and apparatus realize targeted process monitoring and also can obtain more monitoring data, thereby improving the efficiency for monitoring the performance of the virtual machine.

Description

technical field [0001] The invention relates to computer technology, in particular to a virtual machine monitoring method and device. Background technique [0002] Existing virtual machine monitoring technology generally uses some monitoring software (such as Top monitoring tool) to monitor process CPU (Central Processing Unit, central processing unit), internal memory, I / O (Input / Output, input / output) in the virtual machine. Usage is monitored. However, such a monitoring method cannot monitor the detailed operation status of the business process in the system, such as: the type of system call invoked by the business process, the time spent, the number of failures, and the like. Moreover, the monitoring data obtained by existing monitoring methods cannot make effective judgments on the performance of business processes in the system. For example, when the CPU load of the system is too high, if only the process CPU, memory , I / O (Input / Output, input / output) usage data, it i...

AI Technical Summary

Problems solved by technology

However, such a monitoring method cannot monitor the detailed operating status of the business process in the system, such as: the type of system call invoked by the business process, the time spent, the number of failures, etc.

Moreover, the monitoring data obtained by existing monitoring methods cannot make effective judgments on the performance of business processes in the system. For example, when the CPU load of the system is too high, if only the process CPU, memory , I / O (Input / Output, input / output) usage data, it is impossible to analyze which system call of which process caused the system CPU load to be too high

Therefore, the monitoring content of the existing virtual machine monitoring technology is limited, and the monitoring efficiency is not high

Images