Abstract
In a commercial cloud environment, exhaustive resource profiling is beneficial to cope with the intrinsic variability of the virtualised environment, allowing to promptly identify performance degradation. In the context of its commercial cloud initiatives, CERN has acquired extensive experience in benchmarking commercial cloud resources. Ultimately, this activity provides information on the actual delivered performance of invoiced resources. In this report we discuss the experience acquired and the results collected using several fast benchmark applications adopted by the HEP community. These benchmarks span from open-source benchmarks to specific user applications and synthetic benchmarks. The workflow put in place to collect and analyse performance metrics is also described.
Highlights
Since the early days of computing, benchmarking has represented a powerful method to assess the relative performance of computing systems
The benchmark widely adopted by the High Energy Physics (HEP) community to measure CPU performance is HEP-SPEC06 (HS06) [1]
The results show that simultaneous multithreading (SMT) enabled implies a benefit of 24% in throughput for HS06 when all hardware threads are in use
Summary
- Benchmarking and accounting for the (private) cloud J Belleman and U Schwickerath. - Benchmarking Geant for spallation neutron source calculations Douglas D. - Benchmarking and accounting for the (private) cloud J Belleman and U Schwickerath. - Benchmarking Geant for spallation neutron source calculations Douglas D. DiJulio, Konstantin Batkov, John Stenander et al. - Benchmarking of Advanced Control Strategies for a Simulated Hydroelectric System S. This content was downloaded from IP address 129.13.72.197 on 23/11/2017 at 15:18. M Alef, C Cordeiro, A De Salvo, A Di Girolamo, L Field, D Giordano, M Guerri, F C Schiavi and A Wiebalck
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
