Approximating warps with intra-warp operand value similarity
- 1 March 2016
- conference paper
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE) in 2016 IEEE International Symposium on High Performance Computer Architecture (HPCA)
Abstract
Value locality, the recurrence of a previously-seen value, has been the enabler of myriad optimization techniques in traditional processors. Value similarity relaxes the constraint of value locality by allowing values to differ in the lowest significant bits where values are micro-architecturally near. With the end of Dennard Scaling and the turn towards massively parallel accelerators, we revisit value similarity in the context of GPUs. We identify a form of value similarity called intra-warp operand value similarity, which is abundant in GPUs. We present Warp Approximation, which leverages intra-warp operand value similarity to trade off accuracy for energy. Warp Approximation dynamically identifies intra-warp operand value similarity in hardware, and executes a single representative thread on behalf of all the active threads in a warp, thereby producing a representative value with approximate value locality. This representative value can then be stored compactly in the register file as a value similar scalar, reducing the read and write energy when dealing with approximate data. With Warp Approximation, we can reduce execution unit energy by 37%, register file energy by 28%, and improve overall GPGPU energy efficiency by 26% with minimal quality degradation.Keywords
This publication has 34 references indexed in Scilit:
- RumbaPublished by Association for Computing Machinery (ACM) ,2015
- Warped-compressionPublished by Association for Computing Machinery (ACM) ,2015
- SAGEPublished by Association for Computing Machinery (ACM) ,2013
- Quality programmable vector processors for approximate computingPublished by Association for Computing Machinery (ACM) ,2013
- Exploiting uniform vector instructions for GPGPU performance, energy efficiency, and opportunistic reliability enhancementPublished by Association for Computing Machinery (ACM) ,2013
- Power-efficient computing for compute-intensive GPGPU applicationsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2013
- Architecture support for disciplined approximate programmingPublished by Association for Computing Machinery (ACM) ,2012
- Improving GPU performance via large warps and two-level warp schedulingPublished by Association for Computing Machinery (ACM) ,2011
- Managing performance vs. accuracy trade-offs with loop perforationPublished by Association for Computing Machinery (ACM) ,2011
- Fuzzy Memoization for Floating-Point Multimedia ApplicationsInternational Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2005