Average Delay Guarantee in Server Systems Using Admission Control

Abstract Although server technology provides a mean to support a wide range of online services and applications, their ad-hoc configuration poses significant challenges to the performance, availability and economical costs of applications. One of the main concerns is that under a heavy load, the delay introduced by a server in the transaction process may grow unbounded. This paper precisely addresses this issue. First, we present the design of a server model as a non-linear continuous-time model. Second, we develop an admission control algorithm that allow to ensure a maximum average delay on the server. Model and control algorithm were implemented and applied on the standard PostgreSQL database server running the TPC-C warehouse application. The experiments show that the proposed method provides significant benefits for database servers management.

[1]  J. Little A Proof for the Queuing Formula: L = λW , 1961 .

[2]  Martin Arlitt,et al.  A workload characterization study of the 1998 World Cup Web site , 2000, IEEE Netw..

[3]  Aurel A. Lazar,et al.  Joint scheduling and admission control for ATS-based switching nodes , 1992, SIGCOMM '92.

[4]  Shang Zhi,et al.  A proof of the queueing formula: L=λW , 2001 .

[5]  Yixin Diao,et al.  Using MIMO feedback control to enforce policies for interrelated metrics with application to the Apache Web server , 2002, NOMS 2002. IEEE/IFIP Network Operations and Management Symposium. ' Management Solutions for the New Communications World'(Cat. No.02CH37327).

[6]  Erich M. Nahum,et al.  A method for transparent admission control and request scheduling in e-commerce web sites , 2004, WWW '04.

[7]  Malur K. Sundareshan,et al.  Numerical Methods for Modeling Computer Networks Under Nonstationary Conditions , 1990, IEEE J. Sel. Areas Commun..

[8]  Martin F. Arlitt,et al.  Web server workload characterization: the search for invariants , 1996, SIGMETRICS '96.

[9]  Adam Wierman,et al.  How to Determine a Good Multi-Programming Level for External Scheduling , 2006, 22nd International Conference on Data Engineering (ICDE'06).

[10]  Yixin Diao,et al.  Feedback Control of Computing Systems , 2004 .

[11]  Maria Kihl,et al.  Analysis of admission control mechanisms using non-linear control theory , 2003, Proceedings of the Eighth IEEE Symposium on Computers and Communications. ISCC 2003.

[12]  Dennis Shasha Optimizing database performance , 1993 .

[13]  A. Robertsson,et al.  Admission control for Web server systems - design and experimental evaluation , 2004, 2004 43rd IEEE Conference on Decision and Control (CDC) (IEEE Cat. No.04CH37601).

[14]  Hans-Ulrich Heiß,et al.  Adaptive Load Control in Transaction Processing Systems , 1991, VLDB.

[15]  Sujata Banerjee,et al.  A simple approximation for modeling nonstationary queues , 1996, Proceedings of IEEE INFOCOM '96. Conference on Computer Communications.

[16]  Ronald C. Dodge,et al.  Preserving QoS of e-commerce sites through self-tuning: a performance model approach , 2001, EC '01.

[17]  Joseph L. Hellerstein,et al.  Using Control Theory to Achieve Service Level Objectives In Performance Management , 2002, Real-Time Systems.

[18]  Martin Arlitt,et al.  Workload Characterization of the 1998 World Cup Web Site , 1999 .

[19]  林偉元 電腦叢集上QuickTime Streaming Server研究 , 2003 .