Approximate computation with outlier detection in Topaz
Name
Rinard_Approximate computation.pdf
Size
2.97 MB
Format
Adobe PDF
Checksum (MD5)
bfb35c25cceeb7ddfe1b5bb10768cd1f
Author(s)
Rinard, Martin C
Achour, Sara
Date Issued
October 2015
Journal
Proceedings of the 2015 ACM SIGPLAN International Conference on Object-Oriented Programming, Systems, Languages, and Applications - OOPSLA 2015
Publisher
Association for Computing Machinery
Citation
Achour, Sara, and Martin C. Rinard. "Approximate Computation with Outlier Detection in Topaz." Proceedings of the 2015 ACM SIGPLAN International Conference on Object-Oriented Programming, Systems, Languages, and Applications - OOPSLA 2015, 25-30 October, 2015, Pittsburgh, Pennsylvania, ACM Press, 2015, pp. 711–30.
Version
Author's final manuscript
Abstract
We present Topaz, a new task-based language for computations that execute on approximate computing platforms that may occasionally produce arbitrarily inaccurate results. Topaz maps tasks onto the approximate hardware and integrates the generated results into the main computation. To prevent unacceptably inaccurate task results from corrupting the main computation, Topaz deploys a novel outlier detection mechanism that recognizes and precisely reexecutes outlier tasks. Outlier detection enables Topaz to work effectively with approximate hardware platforms that have complex fault characteristics, including platforms with bit pattern dependent faults (in which the presence of faults may depend on values stored in adjacent memory cells). Our experimental results show that, for our set of benchmark applications, outlier detection enables Topaz to deliver acceptably accurate results (less than 1% error) on our target approximate hardware platforms. Depending on the application and the hardware platform, the overall energy savings range from 5 to 13 percent. Without outlier detection, only one of the applications produces acceptably accurate results.
MIT Department
Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
Terms of Use
Creative Commons Attribution-Noncommercial-Share Alike
Persistent DSpace Link
DOI of Published Version
https://doi.org/10.1145/2814270.2814314
