CISUC

Dependability Benchmarks for OLTP Systems

Authors

Abstract

On-Line Transaction Processing (OLTP) systems are the core of the information systems used to support the daily operations of organizations. These systems, whose correct behavior is of extreme importance, use a database management system (DBMS) that assures transactional properties and provides data recovery features. Although these systems represent the best examples of business-critical systems, no practical way has been pro-posed so far to characterize and compare alternative solutions concerning dependability features. In fact, transac-tional systems industry holds a well established infrastructure for performance evaluation, where the benchmarks developed by the Transactional Performance Processing Council (TPC) represent the most important bench-marking initiative of all computer industry. However, as performance evaluation has been the main focus, de-pendability benchmarking of transactional systems has been largely disregarded over the years. This chapter proposes a dependability benchmark for OLTP application environments that specifies the measures and all components required to evaluate both the performance and key dependability features of OLTP systems, with particular emphasis on availability. The benchmark is presented through a concrete example of benchmarking the performance and dependability of several different transactional systems configurations. The results obtained show clearly that it is possible to apply dependability benchmarking in transactional systems in a very successful way.

Subject

Dependability benchmarking

Related Project

DBench - Dependability Benchmarking

Book Chapter

Dependability Benchmarking for Computer Systems (ISBN: 978-0-470-23055-8), Wiley-IEEE Computer Society Press, August 2008

Cited by

Year 2010 : 1 citations

 1. Anakreon Mentis, Panagiotis Katsaros, Lefteris Angelis, George Kakarontzas, “Quantification of interacting runtime qualities in software architectures: insights from transaction processing in client-server architectures”, Information and Software Technology, Volume 52, Issue 12, December 2010.