Design and configuration of distributed job processing systems

A key criterion in the design, procurement and use of computer systems is performance. Performance typically means the throughput and response time of a system. The effects of poorly performing systems range from dissatisfied users to high penalties for companies due to missed processing deadlines. As a result of continuously increasing hardware performance, companies often solve performance problems by replacing existing hardware with faster machines. One consequence can be that they achieve a performance increase, since, for example, the processors themselves are faster, but the overall performance increase is less than expected. The reason for this is that the combination of hardware and software does not match. For system designers it would be helpful to have a systematic method which supports them in the design of new systems and in the extension of existing systems. The need for a systematic configuration method is motivated by a typical example application from the B2B world. In the financial industry, banks have to deal with several payment messages standards like EDIFACT [Uni97] or S.W.I.F.T. [SWI97]. Such messages have to be converted into the banks' internal representation for further processing. Such converters have to handle message size ranging from some 100 bytes to about 60 MB and have to fulfil certain performance requirements such as throughput or meeting deadlines. To achieve the performance goals, identification of the hardware and software configuration is an important step in the implementation of a distributed message converter system. The problem of sizing a system is a topic of the research field of Capacity Planning. Menascé et al. [MA02] have developed a general methodology for system capactity planning. Typically this requires that several configurations are tried, analysed and adapted to find a suitable solution. The experience of the system designer plays an important role here in the identification of suitable solutions. But to date hardly any systematic approaches are known for the configuration of systems. This thesis presents a systematic approach for the cost performance analysis of distributed job processing systems based on given requirements on throughput and system response time. Our method allows us to search for suitable configurations while minimizing the use of expen-iii iv Abstract sive methods for performance evaluation to the largest degree. The method is organized into a hardware and a software configuration step. For each of these configuration steps algorithms were developed. These are based on queuing network models to predict system performance. …