CapSys 2.0 – Visual Capacity Planning for IBM Mainframe Systems

Conference: MMB 2008 - 14th GI/ITG Conference - Measurement, Modelling and Evalutation of Computer and Communication Systems
03/31/2008 - 04/02/2008 at Dortmund, Germany

Proceedings: MMB 2008

Pages: 3Language: englishTyp: PDF

Personal VDE Members are entitled to a 10% discount on this title

Tilev, Milen; Müller-Clostermann, Bruno (University of Duisburg-Essen, Institute for Computer Science and Business Information, Systems (ICB), Schützenbahn 70, D-45117 Essen, Germany)

CapSys 2.0 supports the daily capacity planning practice based on well-known techniques and guidelines. The tool works with workload specific utilisation data from real systems, which is permanently measured and stored in a reporting database. For planning and analysis purposes, CapSys uses simple models of recent mainframe hardware. Furthermore, it explores the various interrelationships between different mainframe workload types, special constructs like sysplexes and IRD clusters, and hardware components like so called specialty engines and processor pools (for all mainframe-related notions cf. [1]). A workload within CapSys’ scope is defined as a time series of utilisation data caused by a LPAR (Logical PARtition) on a certain type of processor pool (e.g. general, Linux, Java, coupling facility, etc.).