Check out the new USENIX Web site.

Home About USENIX Events Membership Publications Students
OSDI '02 Paper    [OSDI '02 Tech Program Index]

Pp. 225-238 of the Proceedings
 

Integrated Resource Management for Cluster-based Internet Services

Kai Shen*Hong Tang+Tao Yang+$Lingkun Chu+
kshen@cs.rochester.eduhtang@cs.ucsb.edutyang@cs.ucsb.edulkchu@cs.ucsb.edu

 

*Dept. of Computer Science, University of Rochester, Rochester, NY 14627
+Dept. of Computer Science, University of California, Santa Barbara, CA 93106
$Ask Jeeves/Teoma Technologies, Piscataway, NJ 08854

Abstract

Client request rates for Internet services tend to be bursty and thus it is important to maintain efficient resource utilization under a wide range of load conditions. Network service clients typically seek services interactively and maintaining reasonable response time is often imperative for such services. In addition, providing differentiated service qualities and resource allocation to multiple service classes can also be desirable at times. This paper presents an integrated resource management framework (part of Neptune system) that provides flexible service quality specification, efficient resource utilization, and service differentiation for cluster-based services. This framework introduces the metric of quality-aware service yield to combine the overall system efficiency and individual service response time in one flexible model. Resources are managed through a two-level request distribution and scheduling scheme. At the cluster level, a fully decentralized request distribution architecture is employed to achieve high scalability and availability. Inside each service node, an adaptive scheduling policy maintains efficient resource utilization under a wide range of load conditions. Our trace-driven evaluations demonstrate the performance, scalability, and service differentiation achieved by the proposed techniques.


1   Introduction

Previous studies show that the client request rates for Internet services tend to be bursty and fluctuate dramatically [5,10,11]. For example, the daily peak-to-average load ratio at Internet search service Ask Jeeves (www.ask.com) is typically 3:1 and it can be much high