Computer (Long Beach, Calif.), 2016-10, Vol.49 (10), p.30-37
2016
Volltextzugriff (PDF)
Details
- Autor(en) / Beteiligte
- Titel
- Power, Reliability, and Performance: One System to Rule them All
- Ist Teil von
- Computer (Long Beach, Calif.), 2016-10, Vol.49 (10), p.30-37
- Ort / Verlag
- New York: IEEE
- Erscheinungsjahr
- 2016
- Quelle
- IEEE Electronic Library Online
- Beschreibungen/Notizen
- In a design based on the Charm++ parallel programming framework, an adaptive runtime system dynamically interacts with a datacenter's resource manager to control power by intelligently scheduling jobs, reallocating resources, and reconfiguring hardware. It simultaneously manages reliability by cooling the system to the running application's optimal level and maintains performance through load balancing.
- Sprache
- Englisch
- Identifikatoren
- ISSN: 0018-9162eISSN: 1558-0814DOI: 10.1109/MC.2016.310Titel-ID: cdi_ieee_primary_7598172
- Format
- –
- Schlagworte
- Adaptive control systems, Adaptive systems, Charm, Computer networks, Computer programming, Design engineering, Dynamic scheduling, Dynamical systems, Energy efficiency, energy-aware systems, energy-efficient computing, Hardware, High performance computing, HPC, Load management, Optimization, parallel systems, power management, Power system management, Program processors, Reliability, Resource scheduling, Running, Runtime, Temperature measurement, temperature-aware design