Duality Cache for Data Parallel Acceleration
Duality Cache for Data Parallel Acceleration
复制标题
DOI：
10.1145/3307650.3322257
复制
发表时间：
2019-06
期刊：
影响因子：
--
通讯作者：
Daichi Fujiki;S. Mahlke;R. Das
中科院分区：
文献类型：
--
作者：
Daichi Fujiki;S. Mahlke;R. Das
Duality Cache是一种高速缓存内计算架构，使通用数据并行应用程序能够在高速缓存上运行。本文提出了一种构建DualityCache系统栈的整体方法，该方法采用了执行缓存内浮点运算和超越函数的技术，实现了数据并行执行模型，设计了一个接受现有CUDA程序的编译器，并提供了适应各种工作负载特性的灵活性。Duality Cache架构中存在的大规模并行性使GPU基准测试的性能提高了3.6倍，OpenACC基准测试的性能提高了4.0倍。重新利用现有缓存可为CPU提供72.6倍的性能，而面积成本仅为3.5%。Duality Cache的能耗比GPU低5.8倍，比CPU低21倍。
Duality Cache is an in-cache computation architecture that enables general purpose data parallel applications to run on caches. This paper presents a holistic approach of building Duality Cache system stack with techniques of performing in- cache floating point arithmetic and transcendental functions, enabling a data-parallel execution model, designing a compiler that accepts existing CUDA programs, and providing flexibility in adopting for various workload characteristics. Exposure to massive parallelism that exists in the Duality Cache architecture improves performance of GPU benchmarks by 3.6x and OpenACC benchmarks by 4.0x over a server class GPU. Re-purposing existing caches provides 72.6x better performance for CPUs with only 3.5% of area cost. Duality Cache reduces energy by 5.8x over GPUs and 21x over CPUs.