Publikationsansicht

Storage device performance prediction with CART models (2004)

Abstract
Storage device performance prediction is a key element of self-managed storage systems and application planning tasks, such as data assignment. This work explores the application of a machine learning tool, CART models, to storage device modeling. Our approach predicts a device’s performance as a function of input workloads, requiring no knowledge of the device internals. We propose two uses of CART models: one that predicts per-request response times (and then derives aggregate values) and one that predicts aggregate values directly from workload characteristics. After being trained on our experimental platforms, both provide accurate black-box models across a range of test traces from real environments. Experiments show that these models predict the average and 90th percentile response time with an relative error as low as 16%, when the training workloads are similar to the testing workloads, and interpolate well across different workloads. Acknowledgements: We thank the members and companies of the PDL Consortium (including EMC, Hewlett-Packard, Hitachi, Hitachi

Details der Publikation
Download http://citeseerx.ist.psu.edu/viewdoc/summary?doi=?doi=10.1.1.1.2254
Quelle http://www.pdl.cmu.edu/PDL-FTP/Workload/CMU-PDL-04-103.pdf
Mitarbeiter CiteSeerX
Archiv CiteSeerX - Scientific Literature Digital Library and Search Engine (United States)
Keywords Performance prediction, storage device modeling
Typ text
Sprache Englisch
Verknüpfungen 10.1.1.46.3223, 10.1.1.117.3731, 10.1.1.117.39, 10.1.1.42.2080, 10.1.1.111.8732, 10.1.1.20.7675, 10.1.1.12.5287, 10.1.1.2.9355, 10.1.1.29.478, 10.1.1.87.6833, 10.1.1.60.8160, 10.1.1.126.2258, 10.1.1.121.3572, 10.1.1.131.4621, 10.1.1.85.8116, 10.1.1.102.9673, 10.1.1.103.763, 10.1.1.68.3876, 10.1.1.102.1087, 10.1.1.72.3158, 10.1.1.87.8967, 10.1.1.112.7647, 10.1.1.115.5919