Method and system for calculating energy consumption of cloud server based on component energy consumption model

Abstract

The invention discloses a part energy consumption model-based cloud server energy consumption calculation method and system. The method comprises the steps that a hardware detection module in the system collects server hardware information; a model matching module performs matching with a built-in hardware model database according to the obtained hardware information, finds corresponding energy consumption parameters and writes the parameters in an energy consumption model configuration file; a data collection module collects a resource utilization rate; a power consumption estimation module performs system or process real-time energy consumption calculation by using energy consumption models of corresponding parts according to the information obtained by adaptation and the resource utilization rate; the system performs persistent storage in two modes on the calculated server energy consumption information; and the system can transmit the server hardware information and historical andreal-time energy consumption data to a statistics end program. Through the method and the system, a user can view and monitor energy consumption information of a current server in real time; and an energy consumption reference is provided for load scheduling of a data center.

Description

technical field [0001] The invention relates to a method and system for calculating energy consumption of a cloud server, in particular to a method and system for calculating energy consumption of a cloud server based on a component energy consumption model. Background technique [0002] According to statistics, in 2011, the number of data centers worldwide has exceeded 500,000, and their power consumption accounts for 1.5% of the world's total power consumption. Under the general trend of energy saving, establishing a sound energy consumption monitoring mechanism in the data center is the premise of realizing energy planning and energy consumption management. Emerson's survey results in North America show that 51% of data center leaders list monitoring and management capabilities as one of their top three concerns. It can be seen that how to establish effective energy consumption monitoring has become a challenge that CSPs are currently facing. In the context of the ever-...

AI Technical Summary

Problems solved by technology

This method is feasible in a small-scale or homogeneous data center, but it is not feasible in a large-scale, heterogeneous cluster environment, because the installation of external instruments for a large number of servers is expensive and very inconvenient. Easy to manage, while specific energy data acquisition (DAQ) systems (such as IBM Active Energy Manager) are generally platform dependent

Images