通过架构-架构协同设计加速定制架构上的BLAS
Accelerating BLAS on Custom Architecture through Algorithm-Architecture Co-design
摘要:
基本线性代数子程序在高性能和科学计算应用中起着关键作用。在实验上,过去的多核和通用图形处理单元(GPGPU)能够在65 W到240 W的功率下分别实现高达理论峰值性能的15%到57%,用于计算边界运算,如双/单精度的通用矩阵乘法(XGEMM)。对于带宽受限的操作,如单/双精度的矩阵向量乘法(XGEMV),性能仅为多核和GPGPU理论峰值性能的5%至7%。实现BLAS的性能需要摆脱传统的智慧,并通过算法架构协同设计,朝着为BLAS量身定制的定制加速器发展。在本文中,我们提出了加速1级(向量运算),2级(矩阵向量运算),3级(矩阵矩阵运算)BLAS通过算法架构协同设计的粗粒度可重构架构(CGRA)。我们选择REDEFINE CGRA作为我们实验的平台,因为REDEFINE可以通过定制的自定义功能单元(CFU)来支持感兴趣的领域。为了有效地实现BLAS的顺序,我们提出了一个处理单元(PE)的设计,并在PE中执行微架构增强,以实现高达74%的理论峰值性能的PE在DGEMM,40%在DGEMV和20%在双精度的内积(DDOT)。我们将此PE作为CFU附加到REDEFINE CGRA,并展示了我们解决方案的可扩展性。最后,我们展示了PE的性能比商用英特尔微架构、ClearSpeed CSX 700、FPGA和Nvidia GPGPU提高了3- 140倍。
英文摘要:
Basic Linear Algebra Subprograms (BLAS) play key role in high performance and scientific computing applications. Experimentally, yesteryear multicore and General Purpose Graphics Processing Units (GPGPUs) are capable of achieving up to 15 to 57% of the theoretical peak performance at 65W to 240W respectively for compute bound operations like Double/Single Precision General Matrix Multiplication (XGEMM). For bandwidth bound operations like Single/Double precision Matrix-vector Multiplication (XGEMV) the performance is merely 5 to 7% of the theoretical peak performance in multicores and GPGPUs respectively. Achieving performance in BLAS requires moving away from conventional wisdom and evolving towards customized accelerator tailored for BLAS through algorithm-architecture co-design. In this paper, we present acceleration of Level-1 (vector operations), Level-2 (matrix-vector operations), and Level-3 (matrix-matrix operations) BLAS through algorithm architecture co-design on a Coarse-grained Reconfigurable Architecture (CGRA). We choose REDEFINE CGRA as a platform for our experiments since REDEFINE can be adapted to support domain of interest through tailor-made Custom Function Units (CFUs). For efficient sequential realization of BLAS, we present design of a Processing Element (PE) and perform micro-architectural enhancements in the PE to achieve up-to 74% of the theoretical peak performance of PE in DGEMM, 40% in DGEMV and 20% in double precision inner product (DDOT). We attach this PE to REDEFINE CGRA as a CFU and show the scalability of our solution. Finally, we show performance improvement of 3-140x in PE over commercially available Intel micro-architectures, ClearSpeed CSX700, FPGA, and Nvidia GPGPUs.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。