Adaptive Partitioning for Irregular Applications on Heterogeneous CPU-GPU Chips
Adaptive Partitioning for Irregular Applications on Heterogeneous CPU-GPU Chips
复制标题
DOI：
10.1016/j.procs.2015.05.213
复制
发表时间：
2015
期刊：
影响因子：
--
通讯作者：
A. Vilches;R. Asenjo;A. Navarro;F. Corbera;Rubén Gran Tejero;M. Garzarán
中科院分区：
文献类型：
--
作者：
A. Vilches;R. Asenjo;A. Navarro;F. Corbera;Rubén Gran Tejero;M. Garzarán
商用处理器由多个CPU内核和一个集成GPU组成。为了充分利用这种类型的架构，需要自动确定如何在两个设备之间划分工作负载。这对于不规则的工作负载来说尤其具有挑战性，因为每次迭代的工作都依赖于数据，并且显示出控制和内存的差异。本文提出了一种专门针对在异构CPU-GPU芯片上运行的不规则应用程序设计的新型自适应分区策略。这项工作的主要新奇在于，分配给GPU和CPU的工作负载的大小动态地适应，以最大限度地提高GPU和CPU的利用率，同时平衡设备之间的工作负载。我们的英特尔Haswell架构上使用一组不规则的基准测试的实验结果表明，我们的方法优于详尽的静态和自适应的最先进的方法在性能和能耗方面。
Commodity processors are comprised of several CPU cores and one integrated GPU. To fully exploit this type of architectures, one needs to automatically determine how to partition the workload between both devices. This is specially challenging for irregular workloads, where each iteration's work is data dependent and shows control and memory divergence. In this paper, we present a novel adaptive partitioning strategy specially designed for irregular applications running on heterogeneous CPU-GPU chips. The main novelty of this work is that the size of the workload assigned to the GPU and CPU adapts dynamically to maximize the GPU and CPU utilization while balancing the workload among the devices. Our experimental results on an Intel Haswell architecture using a set of irregular benchmarks show that our approach outperforms exhaustive static and adaptive state-of-the-art approaches in terms of performance and energy consumption.