On Allocation Policies for Power and Performance

TL;DR

This paper investigates dynamic server allocation policies aimed at maximizing data center revenue while minimizing power usage, using Wikipedia workload traces to evaluate effectiveness under non-stationary conditions.

Contribution

It introduces novel dynamic allocation schemes that adapt to workload predictions and compensate for forecasting errors to improve power efficiency and revenue.

Findings

Proposed schemes perform well with non-stationary workloads.

Forecast error compensation improves resource allocation accuracy.

Experimental results validate the effectiveness of the policies.

Abstract

With the increasing popularity of Internet-based services and applications, power efficiency is becoming a major concern for data center operators, as high electricity consumption not only increases greenhouse gas emissions, but also increases the cost of running the server farm itself. In this paper we address the problem of maximizing the revenue of a service provider by means of dynamic allocation policies that run the minimum amount of servers necessary to meet user's requirements in terms of performance. The results of several experiments executed using Wikipedia traces are described, showing that the proposed schemes work well, even if the workload is non-stationary. Since any resource allocation policy requires the use of forecasting mechanisms, various schemes allowing compensating errors in the load forecasts are presented and evaluated.