Anzeige
Erschienen in:
2016 | OriginalPaper | Buchkapitel
Profiling the Performance of Virtualized Databases with the TPCx-V Benchmark
verfasst von : Andrew Bond, Doug Johnson, Greg Kopczynski, H. Reza Taheri
Erschienen in: Performance Evaluation and Benchmarking: Traditional to Big Data to Internet of Things
Verlag: Springer International Publishing
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Abstract
The proliferation of virtualized servers in data centers has conquered the last frontier of bare-iron servers: back-end databases. The multi-tenancy issues of elasticity, capacity planning, and load variation in cloud data centers now coincide with the heavy demands of database workloads; which in turn creates a call for a benchmark specifically intended for this environment.
The TPC–V benchmark will fill this need with a publicly-available, end-to-end benchmark kit. Using a prototype of the kit, we profiled the performance of a server running 60 virtual machines with 48 databases of different sizes, load levels, and workloads. We will show that virtualized servers can indeed handle the elasticity and multi-tenancy requirements of the cloud, but only after careful tuning of the system configuration to avoid bottlenecks.
In this paper, we will provide a brief description of the benchmark, discuss the results and the conclusions drawn from the experiments, and propose future directions for analyzing the performance of cloud data centers by augmenting the capabilities of the TPCx-V benchmark kit.