Exploring the tradeoffs between programmability and efficiency in data-parallel accelerators

Yunsup Lee, Rimas Avizienis, Alex Bishara, Richard Xia, Derek Lockhart, Christopher Batten, Krste Asanović
2011 Proceeding of the 38th annual international symposium on Computer architecture - ISCA '11  
We present a taxonomy and modular implementation approach for data-parallel accelerators, including the MIMD, vector-SIMD, subword-SIMD, SIMT, and vector-thread (VT) architectural design patterns. We introduce Maven, a new VT microarchitecture based on the traditional vector-SIMD microarchitecture, that is considerably simpler to implement and easier to program than previous VT designs. Using an extensive design-space exploration of full VLSI implementations of many accelerator design points,
more » ... evaluate the varying tradeoffs between programmability and implementation efficiency among the MIMD, vector-SIMD, and VT patterns on a workload of compiled microbenchmarks and application kernels. We find the vector cores provide greater efficiency than the MIMD cores, even on fairly irregular kernels. Our results suggest that the Maven VT microarchitecture is superior to the traditional vector-SIMD architecture, providing both greater efficiency and easier programmability.
doi:10.1145/2000064.2000080 dblp:conf/isca/LeeABXLBA11 fatcat:wwaa7pxzhffkdpkxtglil3lu3y