Instantaneous Load Dependent Servers (iLDS) Model for Web Services

In this paper we propose a technique for semiautomatic building of performance models of web service platforms, which we model as instantaneous Load Dependent Servers (iLDS): servers whose service time depends on the workload at a given time. The approach we propose, adopts a black box measurement technique, this means that we can just deploy a service, but we have no access to the underlying server platform configuration both for software and hardware layers. The resulting model can be used by final user or service developers to predict the service behaviour respect to a known workload or to compare different web service platforms. The proposed approach was validated on a simple case study and the measurements put in evidence an interesting result about the common behavior of instantaneous Load Dependent Servers, which can be of general use for modeling this class of systems.

[1]  Raj Jain,et al.  The art of computer systems performance analysis - techniques for experimental design, measurement, simulation, and modeling , 1991, Wiley professional computing.

[2]  Shikharesh Majumdar,et al.  The Stochastic Rendezvous Network Model for Performance of Synchronous Client-Server-like Distributed Software , 1995, IEEE Trans. Computers.

[3]  Ramón Puigjaner,et al.  Using Load Dependent Servers to Reduce the Complexity of Large Client-Server Simulation Models , 2001, Performance Engineering.

[4]  Vipul Mathur,et al.  A computational complexity-aware model for performance analysis of software servers , 2004, The IEEE Computer Society's 12th Annual International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunications Systems, 2004. (MASCOTS 2004). Proceedings..

[5]  Guy Pujolle,et al.  Analysis of a queueing network model with class dependent window flow control , 1992, [Proceedings] IEEE INFOCOM '92: The Conference on Computer Communications.

[6]  Beniamino Di Martino,et al.  Cluster systems and simulation: from benchmarking to off‐line performance prediction , 2007, Concurr. Comput. Pract. Exp..

[7]  Dorina C. Petriu,et al.  The Future of Software Performance Engineering , 2007, Future of Software Engineering (FOSE '07).

[8]  K. Shin,et al.  Performance Guarantees for Web Server End-Systems: A Control-Theoretical Approach , 2002, IEEE Trans. Parallel Distributed Syst..

[9]  C. Murray Woodside.  The Relationship of Performance Models to Data , 2008, SIPEW.

[10]  C. Murray Woodside,et al.  A Calibration Framework for Capturing and Calibrating Software Performance Models , 2008, EPEW.

[11]  C. Murray Woodside,et al.  An "Active Server" model for the performance of parallel programs written using rendezvous , 1986, J. Syst. Softw..

[12]  Liam Murphy,et al.  Performance modeling and prediction of enterprise JavaBeans with layered queuing network templates , 2005, SAVCBS '05.

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

[14]  Paola Inverardi,et al.  Model-based performance prediction in software development: a survey , 2004, IEEE Transactions on Software Engineering.

[15]  Magnus Karlsson,et al.  Dynamic Black-Box Performance Model Estimation for Self-Tuning Regulators , 2005, Second International Conference on Autonomic Computing (ICAC'05).

[16]  H. Schwetman.  Csim18 - the simulation engine , 1996, Proceedings Winter Simulation Conference.

[17]  Sang Hyuk Son,et al.  Feedback Control Architecture and Design Methodology for Service Delay Guarantees in Web Servers , 2006, IEEE Transactions on Parallel and Distributed Systems.

[18]  Edward D. Lazowska,et al.  Incorporating load dependent servers in approximate mean value analysis , 1984, SIGMETRICS '84.