We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.AR

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Hardware Architecture

Title: Evaluating Built-in ECC of FPGA on-chip Memories for the Mitigation of Undervolting Faults

Abstract: Voltage underscaling below the nominal level is an effective solution for improving energy efficiency in digital circuits, e.g., Field Programmable Gate Arrays (FPGAs). However, further undervolting below a safe voltage level and without accompanying frequency scaling leads to timing related faults, potentially undermining the energy savings. Through experimental voltage underscaling studies on commercial FPGAs, we observed that the rate of these faults exponentially increases for on-chip memories, or Block RAMs (BRAMs). To mitigate these faults, we evaluated the efficiency of the built-in Error-Correction Code (ECC) and observed that more than 90% of the faults are correctable and further 7% are detectable (but not correctable). This efficiency is the result of the single-bit type of these faults, which are then effectively covered by the Single-Error Correction and Double-Error Detection (SECDED) design of the built-in ECC. Finally, motivated by the above experimental observations, we evaluated an FPGA-based Neural Network (NN) accelerator under low-voltage operations, while built-in ECC is leveraged to mitigate undervolting faults and thus, prevent NN significant accuracy loss. In consequence, we achieve 40% of the BRAM power saving through undervolting below the minimum safe voltage level, with a negligible NN accuracy loss, thanks to the substantial fault coverage by the built-in ECC.
Comments: 6 pages, 2 figures
Subjects: Hardware Architecture (cs.AR); Machine Learning (cs.LG)
DOI: 10.1109/EMPDP.2019.8671543
Cite as: arXiv:1903.12514 [cs.AR]
  (or arXiv:1903.12514v1 [cs.AR] for this version)

Submission history

From: Behzad Salami [view email]
[v1] Fri, 29 Mar 2019 13:28:22 GMT (5323kb)

Link back to: arXiv, form interface, contact.