Differently from OLTP workloads, OLAP workloads are hardly predictable due to their inherently extemporary nature. Besides, obtaining real OLAP workloads by monitoring the queries actually issued in companies and organizations is quite hard. On the other hand, hardware and software benchmarking in the industrial world, as well as comparative evaluation of novel approaches in the research community, both need reference databases and workloads. In this paper we present CubeLoad, a parametric generator of workloads in the form of OLAP sessions, based on a realistic profile-based model. After describing the main features of CubeLoad, we discuss the results of some tests that show how workloads with very different features can be generated.
Stefano Rizzi, Enrico Gallinucci (2014). CubeLoad: a Parametric Generator of Realistic OLAP Workloads. Heidelberg : Springer.
CubeLoad: a Parametric Generator of Realistic OLAP Workloads
RIZZI, STEFANO;GALLINUCCI, ENRICO
2014
Abstract
Differently from OLTP workloads, OLAP workloads are hardly predictable due to their inherently extemporary nature. Besides, obtaining real OLAP workloads by monitoring the queries actually issued in companies and organizations is quite hard. On the other hand, hardware and software benchmarking in the industrial world, as well as comparative evaluation of novel approaches in the research community, both need reference databases and workloads. In this paper we present CubeLoad, a parametric generator of workloads in the form of OLAP sessions, based on a realistic profile-based model. After describing the main features of CubeLoad, we discuss the results of some tests that show how workloads with very different features can be generated.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
