Integrated Risk Analysis for a Commercial Computing Service in Utility Computing

Recent technological advances in Grid computing enable the virtualization and dynamic delivery of computing services on demand to realize utility computing. In utility computing, computing services will always be available to the users whenever the need arises, similar to the availability of real-world utilities, such as electrical power, gas, and water. With this new outsourcing service model, users are able to define their service needs through Service Level Agreements (SLAs) and only have to pay when they use the services. They do not have to invest on or maintain computing infrastructures themselves and are not constrained to specific computing service providers. Thus, a commercial computing service will face two new challenges: (i) what are the objectives or goals it needs to achieve in order to support the utility computing model, and (ii) how to evaluate whether these objectives are achieved or not. To address these two new challenges, this paper first identifies four essential objectives that are required to support the utility computing model: (i) manage wait time for SLA acceptance, (ii) meet SLA requests, (iii) ensure reliability of accepted SLA, and (iv) attain profitability. It then describes two evaluation methods that are simple and intuitive: (i) separate and (ii) integrated risk analysis to analyze the effectiveness of resource management policies in achieving the objectives. Evaluation results based on simulation successfully demonstrate the applicability of separate and integrated risk analysis to assess policies in terms of the objectives. These evaluation results which constitute an a posteriori risk analysis of policies can later be used to generate an a priori risk analysis of policies by identifying possible risks for future utility computing situations.

[1]  Gregory Francis Pfister,et al.  In search of clusters (2nd ed.) , 1998 .

[2]  Dror G. Feitelson,et al.  Utilization, Predictability, Workloads, and User Runtime Estimates in Scheduling the IBM SP2 with Backfilling , 2001, IEEE Trans. Parallel Distributed Syst..

[3]  Rajkumar Buyya,et al.  Integrated Risk Analysis for a Commercial Computing Service , 2007, IPDPS.

[4]  Robert R. Moeller,et al.  COSO Enterprise Risk Management: Understanding the New Integrated ERM Framework , 2007 .

[5]  B. Schneider,et al.  Service Quality: Research Perspectives , 2003 .

[6]  Scott H. Clearwater,et al.  Computation-at-risk: assessing job portfolio management risk on clusters , 2004, 18th International Parallel and Distributed Processing Symposium, 2004. Proceedings..

[7]  Rajkumar Buyya,et al.  Libra: a computational economy‐based job scheduling system for clusters , 2004, Softw. Pract. Exp..

[8]  P. Altena,et al.  In search of clusters , 2007 .

[9]  Dhabaleswar K. Panda,et al.  Towards provision of quality of service guarantees in job scheduling , 2004, 2004 IEEE International Conference on Cluster Computing (IEEE Cat. No.04EX935).

[10]  David E. Irwin,et al.  Balancing risk and reward in a market-based task service , 2004, Proceedings. 13th IEEE International Symposium on High performance Distributed Computing, 2004..

[11]  David E. Culler,et al.  User-Centric Performance Analysis of Market-Based Cluster Batch Schedulers , 2002, 2nd IEEE/ACM International Symposium on Cluster Computing and the Grid (CCGRID'02).

[12]  R. V. van Nieuwpoort,et al.  The Grid 2: Blueprint for a New Computing Infrastructure , 2003 .

[13]  Dan Tsafrir,et al.  A Short Survey of Commercial Cluster Batch Schedulers , 2005 .

[14]  Rajkumar Buyya,et al.  GridSim: a toolkit for the modeling and simulation of distributed resource management and scheduling for Grid computing , 2002, Concurr. Comput. Pract. Exp..

[15]  Zhiwei Xu,et al.  GridIS: an incentive-based grid scheduling , 2005, 19th IEEE International Parallel and Distributed Processing Symposium.

[16]  Rajkumar Buyya,et al.  Constructing A Grid Simulation with Differentiated Network Service Using GridSim , 2005, International Conference on Internet Computing.

[17]  Rajkumar Buyya,et al.  A taxonomy of market‐based resource management systems for utility‐driven cluster computing , 2006, Softw. Pract. Exp..

[18]  Dan Tsafrir,et al.  Modeling User Runtime Estimates , 2005, JSSPP.

[19]  Scott H. Clearwater,et al.  Computation-at-risk: employing the grid for computational risk management , 2004, 2004 IEEE International Conference on Cluster Computing (IEEE Cat. No.04EX935).

[20]  B. Looy,et al.  Service management : an integrated approach , 2013 .

[21]  Ibm Redbooks,et al.  Workload Management With Loadleveler , 2001 .

[22]  Rajkumar Buyya,et al.  Cluster computing: the commodity supercomputer , 1999 .

[23]  David A. Lifka,et al.  The ANL/IBM SP Scheduling System , 1995, JSSPP.

[24]  John Wilkes,et al.  Profitable services in an uncertain world , 2005, ACM/IEEE SC 2005 Conference (SC'05).

[25]  Rajkumar Buyya,et al.  Managing Risk of Inaccurate Runtime Estimates for Deadline Constrained Job Admission Control in Clusters , 2006, 2006 International Conference on Parallel Processing (ICPP'06).

[26]  クリストファー リンキスト.  Utility Computing , 2004, Lecture Notes in Computer Science.

[27]  Gregory F. Pfister,et al.  In Search of Clusters , 1995 .

[28]  M. Crouhy,et al.  The essentials of risk management , 2005 .

[29]  Rajkumar Buyya,et al.  Utility Computing and Global Grids , 2006, ArXiv.

[30]  Rajkumar Buyya,et al.  Pricing for Utility-Driven Resource Management and Allocation in Clusters , 2007, Int. J. High Perform. Comput. Appl..

[31]  Rajkumar Buyya,et al.  On incorporating differentiated levels of network service into GridSim , 2007, Future Gener. Comput. Syst..

[32]  Rajkumar Buyya,et al.  A taxonomy of market-based resource management systems for utility-driven cluster computing , 2006 .