EMEURO: A framework for generating multi-purpose accelerators via deep learning
EMEURO: A framework for generating multi-purpose accelerators via deep learning
复制标题
EMEURO：通过深度学习生成多功能加速器的框架
DOI：
10.1109/cgo.2015.7054193
复制
发表时间：
2015
期刊：
影响因子：
--
通讯作者：
K. Olukotun
中科院分区：
文献类型：
--
作者：
Lawrence C. McAfee;K. Olukotun
近似计算是一种非常有前途的设计范例，可以跨越CPU的功率墙，主要是由牺牲输出质量以获得性能、能量和容错能力的显着提高的潜力所驱动。不幸的是，现有的解决方案主要集中在新的编程模型或新的硬件设计上，在这两端之间为基于软件的优化留下了很大的空间。为了填补这一空白，应该针对编译和运行时阶段进行额外的努力，这对控制许多近似子计算的交互以形成良好优化的应用程序具有关键影响。本文介绍了EMEURO，基于神经网络（NN）的仿真和加速平台。通过重构算法，使其具有与NN相同的数据流，EMEURO能够以最小的设计工作量在多个领域实现显著的加速。EMEURO使用新的基于NN的近似计算技术，包括有效搜索高维子程序空间的方法，以及在运行时对错误的细粒度控制。在近似误差为0.1%-10%的情况下，EMEURO算法的最大加速比为原算法的7×-109×.
Approximate computing is a very promising design paradigm for crossing the CPU power wall, primarily driven by the potential to sacrifice output quality for significant gains in performance, energy, and fault tolerance. Unfortunately, existing solutions have primarily either focused on new programming models, or new hardware designs, leaving significant room between these two ends for software-based optimizations. To fill this void, additional efforts should target the compilation and runtime stages, which have a critical impact on controlling the interactions of the many approximate subcomputations to form a well-optimized application. This paper presents EMEURO, a neural-network (NN) based emulation and acceleration platform. By restructuring algorithms to have the same data flow as a NN, EMEURO is able to achieve significant speedup across several domains with minimal design effort. EMEURO uses novel NN-based approximate computing techniques, including methods for efficiently searching the high-dimension subroutine space, and fine-grain control of error during runtime. EMEURO is able to achieve 7×-109× maximum speedup over the original algorithm with 0.1%-10% approximation error.