Affine Vector Cache for memory bandwidth savings - ENS de Lyon - École normale supérieure de Lyon Access content directly
Reports (Research Report) Year : 2011

Affine Vector Cache for memory bandwidth savings


Preserving memory locality is a major issue in highly-multithreaded architectures such as GPUs. These architectures hide latency by maintaining a large number of threads in flight. As each thread needs to maintain a private working set, all threads collectively put tremendous pressure on on-chip memory arrays, at significant cost in area and power. We show that thread-private data in GPU-like implicit SIMD architectures can be compressed by a factor up to 16 by taking advantage of correlations between values held by different threads. We propose the Affine Vector Cache, a compressed cache design that complements the first level cache. Evaluation by simulation on the SDK and Rodinia benchmarks shows that a 32KB L1 cache assisted by a 16KB AVC presents a 59% larger usable capacity on average compared to a single 48KB L1 cache. It results in a global performance increase of 5.7% along with an energy reduction of 11% for a negligible hardware cost.
Fichier principal
Vignette du fichier
Collange_AffineVectorCache_TR13.pdf (358 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

ensl-00649200 , version 1 (07-12-2011)


  • HAL Id : ensl-00649200 , version 1


Caroline Collange, Alexandre Kouyoumdjian. Affine Vector Cache for memory bandwidth savings. [Research Report] ENS de Lyon. 2011. ⟨ensl-00649200⟩
552 View
484 Download


Gmail Facebook Twitter LinkedIn More