Answers are grounded in this paper and cite their sources.
Dynamic frequency and voltage scaling (DVFS) techniques have been widely used for meeting constraints. Single-chip many-core systems bring new challenges owing to the large number operating points and the shift to message passing from shared memory communication. , however, has been mostly studied on single-chip systems with one or few cores, without the impact of the communication among cores. This paper evaluates the impact of and frequency scaling on the performance and power of many-core systems with message (MP) based communication, and proposes a power management policy that leverages the pattern information to efficiently traverse the search space for finding the optimal and frequency operating point. We conduct experiments on a 48-core Intel Single-Chip Computer (SCC), as our target many-core platform. The paper first introduces the runtime infrastructure and the application suite we have designed for an in-depth evaluation the SCC. We then quantify the effects of frequency perturbations on performance and energy . Experimental results show that runtime communication patterns lead to significant differences power/performance tradeoffs in many-core systems with MP-based communication. show that the proposed power management policy achieves up to the 70% energy-delayproduct (EDP) improvements compared to existing DVFS policies, while meeting the performance .
No takes yet. Share an insight, caveat, or question.
Bartolini et al. (2014) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: