A Case for a Flexible Scalar Unit in SIMT Architecture
Authors:
- Yi YangYYYi Yang
- Ping XiangPXPing Xiang
- Michael MantorMMMichael Mantor
- Norman RubinNRNorman Rubin
- Lisa HsuLHLisa Hsu
- Qunfeng DongQDQunfeng Dong
- Huiyang ZhouHZHuiyang Zhou
- 1 May 2014
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 93-102
Abstract
The wide availability and the Single-Instruction Multiple-Thread (SIMT)-style programming model have made graphics processing units (GPUs) a promising choice for high performance computing. However, because of the SIMT style processing, an instruction will be executed in every thread even if the operands are identical for all the threads. To overcome this inefficiency, the AMD's latest Graphics Core Next (GCN) architecture integrates a scalar unit into a SIMT unit. In GCN, both the SIMT unit and the scalar unit share a single SIMT style instruction stream. Depending on its type, an instruction is issued to either a scalar or a SIMT unit. In this paper, we propose to extend the scalar unit so that it can either share the instruction stream with the SIMT unit or execute a separate instruction stream. The program to be executed by the scalar unit is referred to as a scalar program and its purpose is to assist SIMT-unit execution. The scalar programs are either generated from SIMT programs automatically by the compiler or manually developed by expert developers. We make a case for our proposed flexible scalar unit through three collaborative execution paradigms: data prefetching, control divergence elimination, and scalar-workload extraction. Our experimental results show that significant performance gains can be achieved using our proposed approaches compared to the state-of-art SIMT style processing.Keywords
This publication has 26 references indexed in Scilit:
- Warp-level divergence in GPUs: Characterization, impact, and mitigation2014
- Orchestrated scheduling and prefetching for GPGPUs2013
- OWL2013
- Accelerating CUDA graph algorithms at maximum warp2011
- COMPASS2010
- High performance discrete Fourier transforms on graphics processors2008
- Dynamic Warp Formation and Scheduling for Efficient GPU Control Flow2007
- Using a user-level memory thread for correlation prefetching2002
- Out-of-order vector architectures1997
- Spert-II: a vector microprocessor systemComputer, 1996