Virtualized cloud platform energy consumption measurement method and system based on tree regression

Abstract

The invention provides a virtualized cloud platform energy consumption measurement method based on tree regression and a virtualized cloud platform energy consumption measurement system based on the tree regression. The method comprises the following steps: information collection, namely, sending acquired information to assigned control nodes by physical machines, wherein the acquired information comprises the resource information of the physical machines, the resource information of virtual machines, and the energy consumption information obtained from meters; parameter training, namely, the control nodes being responsible for implementing linear fitting on a data set in sections by using a tree regression algorithm to obtain resource energy consumption model parameters alpha and gamma in each subset; energy consumption calculation, namely, implementing calculation by the control nodes according to the fitted parameters and the information of the virtual machines to obtain the energy consumption of the virtual machines. The method and the system have the beneficial effects that as linear models are established in sections for the usage rates of various resources by using a method based on the tree regression, the defect that a traditional single linear model is not high enough in precision is overcome; a regression tree model is simple and efficient to calculate, and reflects a higher efficiency of updating real-time energy consumption model parameters.

Description

technical field [0001] The invention relates to the field of data processing, in particular to a method and system for measuring energy consumption of a virtualized cloud platform based on tree regression. Background technique [0002] With the rapid development of cloud computing, various enterprises have invested a lot of money to build their own data centers. Therefore, the fine-grained monitoring and management of energy consumption is the main technical problem that needs to be solved urgently. Traditional data centers use electric meters to monitor the energy consumption of various devices, but there are still many data center scheduling strategies that rely on real-time acquisition of energy consumption information of servers, applications running on them, and virtual machines. Especially for the server energy consumption of the virtualized cloud platform, the energy consumption of the software running on it, and the energy consumption of the virtual machine running ...

AI Technical Summary

Problems solved by technology

Obtain relevant data for modeling, and do not consider complex multi-machine situations

[0005] Eg: Beijing University of Posts and Telecommunications proposed a virtual machine energy consumption measurement method based on virtual machine internal resource monitoring. The shortcoming is that the resource information displayed inside the virtual machine does not accurately reflect the actual physical resource usage of the virtual machine. It is relatively impractical to run different programs on virtual machines with different configurations, and the accuracy is not high enough

[0013] (1) The energy consumption monitoring of the current data center mainly uses electric meters to monitor the energy consumption of various equipment and servers in the data center, and lacks a model method to fine-grain the energy consumption of the server and the software and virtual machines running on it. monitor

[0014] (2) The installation of an agent program inside the virtual machine to collect the information of the virtual machine breaks the security assumption and affects the privacy of users, which is not feasible for the actual virtual machine service provider

Moreover, the modeling of resource usage and energy consumption inside the virtual machine is not accurate enough

[0015] (3) Establishing a linear model for each virtual machine by dividing physical machine resources to infer virtual machine energy consumption is a better solution at present, but the pure linear model is not accurate enough

[0016] (4) The model of the existing technology is not flexible enough, and it cannot be self-adapted to a specific benchmark. The error of the result measured by the model method will vary with the application program, which is unstable and its accuracy will also vary greatly

[0017] (5) The model parameters faced by different resource utilization rates may be different. The existing linear method adopts the method of separately training parameters in a segmented manner, which is simply segmented in an artificial way, which is not objective enough

[0018] (6) In the nonlinear method adopted, the calculation is complicated, and the model parameters cannot be updated in time for real-time measurement of energy consumption, resulting in large errors

Images