Brief announcement: Integrated resource allocation in heterogeneous SAN data centers

Abstract

Modern data centers are complex distributed environments with application workloads requiring multiple resources like processing (CPU), storage and network. Allocation of these resources to workloads needs to be handled in an integrated manner to adequately capture the relationships between different resource nodes like connectivity between an application server and storage controller in the storage area network (SAN). As data centers grow over time, heterogeneous resources coexist at the same time and this heterogeneity adds further complexity to manual resource allocation. In this work, we describe various challenges and key insights in performing fast, automatic integrated resource allocation. We briefly introduce our novel framework called SPARK (Stable-Proposals-And-Resource Knapsacks) that uses server virtualization to address combined placement of application data and CPU in SAN data centers. SPARK is based on two well-studied problems - Stable Marriage and Knapsacks - and is simple, fast, versatile and highly scalable. Our initial experiments show promise of our approach, consistently outperforming natural candidate algorithms by 30-40% and being within 4% of the LP-based optimal values for a wide range of experiments. Copyright © 2007 ACM.