Modeling the Energy Efficiency of GEMM using Optical Random Access Memory
Modeling the Energy Efficiency of GEMM using Optical Random Access Memory
复制标题
DOI：
10.1109/hpec55821.2022.9926291
复制
发表时间：
2022-09
期刊：
影响因子：
--
通讯作者：
Bingyi Zhang;Akhilesh R. Jaiswal;Clynn Mathew;R. T. Lakkireddy;Ajey P. Jacob;Sasindu Wijeratne;V. Prasanna
中科院分区：
文献类型：
--
作者：
Bingyi Zhang;Akhilesh R. Jaiswal;Clynn Mathew;R. T. Lakkireddy;Ajey P. Jacob;Sasindu Wijeratne;V. Prasanna
一般矩阵-矩阵乘法（GEMM）是许多应用中的关键计算内核。GEMM支持多种硬件平台，包括CPU、GPU、FPGA。为了优化GEMM的性能，开发人员使用片上静电随机存取存储器（E-SRAM）来利用GEMM的数据局部性。然而，为GEMM密集访问E-SRAM会导致大量的能源消耗，这对于商业数据中心来说并不节能。本文对光学静态随机存取存储器（O-SRAM）进行了评价。与传统的E-SRAM相比，O-SRAM具有极低的访问延迟和低的能耗，是一种很有前途的技术。首先，我们提出了一个基于O-SRAM的晶圆级GEMM系统和一个基于E-SRAM的基准系统。其次，我们建立了两种系统的理论性能模型，分析了它们的片上存储器访问能耗。然后，我们进行了基于仿真的实验来评估两种系统的能耗。评估结果表明，基于O-SRAM的系统比基于E-SRAM的基准系统节能7倍。
General matrix-matrix multiplication (GEMM) is the key computation kernel in many applications. GEMM has been supported on various hardware platforms, including CPU, GPU, FPGA. To optimize the performance of GEMM, developers use on-chip electrical static random access memory (E-SRAM) to exploit the data locality of GEMM. However, intensively accessing E-SRAM for GEMM can lead to significant energy consumption, which is not energy-efficient for commercial data centers. In this paper, we evaluate the optical static random access memory (O-SRAM) for GEMM. O-SRAM is a promising tech-nology that has extremely low access latency and low energy consumption compared with the traditional E-SRAM. First, we propose an O-SRAM based wafer-scale system for GEMM and a baseline E-SRAM based system. Second, we build the theoretical performance models of the two systems to analyze their energy consumption of on-chip memory accesses. Then, we conduct simulation-based experiments to evaluate the energy consumption of the two system. The evaluation results show that O-SRAM based system is 7 x more energy efficient than the baseline E-SRAM based system.