A Case Study on Job Scheduling Policy for Workload Characterization and Power Efficiency

TL;DR

This paper analyzes workload and power consumption in real datacenters to identify features that can optimize resource utilization and improve power efficiency.

Contribution

It provides a detailed statistical characterization of datacenter workloads and identifies unique features for optimizing resource use and power efficiency.

Findings

Workload features can be used to optimize resource utilization.

Power consumption correlates with workload characteristics.

Workload patterns are key to improving datacenter energy efficiency.

Abstract

With the increasing popularity of cloud computing, datacenters are becoming more important than ever before. A typical datacenter typically consists of a large number of homogeneous or heterogeneous servers connected by networks. Unfortunately, these servers and network equipment are often under-utilized and power hungry. To improve the utilization of hardware resources and make them power efficiency in datacenters, workload characterization and analysis is at the foundation. In this paper, we characterize and analyze the job arriving rate, arriving time, job length, power consumption, and temperature dissipation in a real world datacenter by using statistical methods. From the characterization, we find unique features in the workload can be used to optimize the resource utilization and power consumption of datacenters