ApproxMA: Approximate memory access for dynamic precision scaling
Refereed conference paper presented and published in conference proceedings

Times Cited
Altmetrics Information

Other information
AbstractMotivated by the inherent error-resilience of emerging recognition, mining, and synthesis (RMS) applications, approximate computing techniques such as precision scaling has been advocated for achieving energy-efficiency gains at the cost of small accuracy loss. Most existing solutions, however, focus on the approximation of on-chip computations without considering that of off-chip data accesses, whose energy consumption may contribute to a significant portion of the total energy. In this work, we propose a novel approximate memory access technique for dynamic precision scaling, namely ApproxMA. To be specific, by taking both runtime data precision constraints and error-resilient capabilities of the application into consideration, ApproxMA determines the precision of data accesses and loads scaled data from off-chip memory for computation. Experimental results with mixture model-based clustering algorithms demonstrate the efficacy of the proposed methodology.
All Author(s) ListTian Y., Zhang Q., Wang T., Yuan F., Xu Q.
Name of Conference25th Great Lakes Symposium on VLSI, GLSVLSI 2015
Start Date of Conference20/05/2015
End Date of Conference22/05/2015
Place of ConferencePittsburgh
Country/Region of ConferenceUnited States of America
Detailed descriptionorganized by ACM,
Volume Number20-22-May-2015
Pages337 - 342
LanguagesEnglish-United Kingdom
KeywordsApproximate computing, Memory access, Precision scaling

Last updated on 2021-17-09 at 23:45