Statistical Modeling of Queuing Sytems for Capacity Management and Planning of It Services

• Activity Number: 420
• Type: Contributed
• Date/Time: Tuesday, July 31, 2012 : 2:00 PM to 3:50 PM
• Sponsor: Quality and Productivity Section
• Author(s): Ta-Hsin Li*+ and Kaan Katircioglu
• Companies: IBM and IBM
• Keywords: Queuing Model ; Estimation ; Application management ; Resourece Allocation ; Workload Consolidation ; Robust

Abstract

In IT services industry, service requests are classified into a plural of categories depending on their nature and priority. The processing of such requests are often governed by service level agreements (SLAs) which specify certain targets for the duration of various processing stages. An important challenge facing the service providers is the management and planning of human resources (including training and consolidation) to satisfy the SLAs while improving the resource utilization and the profitability. The processing of service requires can be regarded as a queuing system in which the requests form a queue and wait to be processed by the available servers who have the required skills. Because of the complexity of the request arrivals, the processing times, and the SLAs, such queuing systems cannot be handled adequately by simple queuing models. In this paper, we discuss some challenges and statistical issues involved in modeling such queuing systems for application management services.