Predictive and Dynamic Resource Allocation for Enterprise Applications

Dynamic resource allocation has the potential to provide significant increases in total revenue in enterprise systems through the reallocation of available resources as the demands on hosted applications change over time. This paper investigates the combination of workload prediction algorithms and switching policies: the former aim to forecast the workload associated with Internet services, the latter switch resources between applications according to certain system criteria. An evaluation of two well known switching policies – the proportional switching policy (PSP) and the bottleneck aware switching policy (BSP) – is conducted in the context of seven workload prediction algorithms. This study uses real-world workload traces consisting of approximately 3.5M requests, and models a multi-tiered, cluster-based, multi-server solution. The results show that a combination of the bottleneck aware switching policy and workload predictions based on an autoregressive, integrated, moving-average model can improve system revenue by as much as 43%.

[1]  Asser N. Tantawi,et al.  Optimal allocation of multiple class resources in computer systems , 1988, SIGMETRICS 1988.

[2]  Asser N. Tantawi,et al.  Optimal Allocation of Multiple Class Resources in Computer Systems , 1988, SIGMETRICS.

[3]  James C. French,et al.  Predicting Indexer Performance in a Distributed Digital Library , 1999, ECDL.

[4]  Stephen A. Jarvis,et al.  Model-driven server allocation in distributed enterprise systems , 2009 .

[5]  Ludmila Cherkasova,et al.  Session-Based Admission Control: A Mechanism for Peak Load Management of Commercial Web Sites , 2002, IEEE Trans. Computers.

[6]  Marin Litoiu,et al.  A performance analysis method for autonomic computing systems , 2007, TAAS.

[7]  Giuseppe Serazzi,et al.  Workload characterization: a survey , 1993, Proc. IEEE.

[8]  Stephen A. Jarvis,et al.  The effect of server reallocation time in dynamic resource allocation , 2009 .

[9]  Giuseppe Serazzi,et al.  Bottlenecks identification in multiclass queueing networks using convex polytopes , 2004, The IEEE Computer Society's 12th Annual International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunications Systems, 2004. (MASCOTS 2004). Proceedings..

[10]  Carey L. Williamson,et al.  Workload Characterization of a Large Systems Conference Web Server , 2009, 2009 Seventh Annual Communication Networks and Services Research Conference.

[11]  Raffaela Mirandola,et al.  Performance Prediction of Web Service Workflows , 2007, QoSA.

[12]  Asser N. Tantawi,et al.  Optimal static load balancing in distributed computer systems , 1985, JACM.

[13]  Daniel A. Menascé.  Workload Characterization , 2003, IEEE Internet Comput..

[14]  Virgílio A. F. Almeida,et al.  Capacity Planning for Web Services: Metrics, Models, and Methods , 2001 .

[15]  Joseph S. Martinich.  Production and Operations Management: An Applied Modern Approach , 1996 .

[16]  Mark S. Squillante,et al.  On maximizing service-level-agreement profits , 2001, EC.

[17]  Daniel A. Menascé,et al.  Scaling for E-Business: Technologies, Models, Performance, and Capacity Planning , 2000 .

[18]  Xiaoyun Zhu,et al.  Statistical service assurances for applications in utility grid environments , 2004, Perform. Evaluation.

[19]  Awi Federgruen,et al.  The Greedy Procedure for Resource Allocation Problems: Necessary and Sufficient Conditions for Optimality , 1986, Oper. Res..

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

[21]  Günter Haring,et al.  Workload modeling for parallel processing systems , 1995, MASCOTS '95. Proceedings of the Third International Workshop on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems.

[22]  F. ArlittMartin,et al.  Web server workload characterization , 1996 .

[23]  D.A. Menasce,et al.  Scaling for e-business , 2000, Proceedings 8th International Symposium on Modeling, Analysis and Simulation of Computer and Telecommunication Systems (Cat. No.PR00728).

[24]  Mark Crovella,et al.  Computer Systems Performance Evaluation , 2007 .

[25]  Giuseppe Serazzi,et al.  Asymptotic Analysis of Multiclass Closed Queueing Networks: Multiple Bottlenecks , 1997, Perform. Evaluation.

[26]  Stephen A. Jarvis,et al.  Partition-based Profit Optimisation for Multi-class Requests in Clusters of Servers , 2007, IEEE International Conference on e-Business Engineering (ICEBE'07).

[27]  Stephen A. Jarvis,et al.  Dynamic Resource Allocation in Enterprise Systems , 2008, 2008 14th IEEE International Conference on Parallel and Distributed Systems.

[28]  Daniel A. Menascé.  Using Performance Models to Dynamically Control E-Business Performance , 2001, MMB.

[29]  Qing Wang,et al.  Workload characterization and customer interaction at e-commerce web servers , 2004 .

[30]  Stephen A. Jarvis,et al.  Predicting the performance of globus monitoring and discovery service (MDS-2) queries , 2003, Proceedings. First Latin American Web Congress.

[31]  Tao Yang,et al.  Selective early request termination for busy internet services , 2006, WWW '06.