Scalable and Fast Lazy Persistency on GPUs
Scalable and Fast Lazy Persistency on GPUs
复制标题
DOI：
10.1109/iiswc50251.2020.00032
复制
发表时间：
2020-10
期刊：
影响因子：
--
通讯作者：
Ardhi Wiratama Baskara Yudha;K. Kimura;Huiyang Zhou;Yan Solihin
中科院分区：
文献类型：
--
作者：
Ardhi Wiratama Baskara Yudha;K. Kimura;Huiyang Zhou;Yan Solihin
GPU应用程序，包括许多科学和机器学习应用程序，越来越需要更大的内存容量。与DRAM相比，NVM有望获得更高的密度和更好的未来扩展潜力。长期运行的GPU应用程序可以通过利用其持久性从NVM中受益，从而允许崩溃恢复内存中的数据。在本文中，我们提出了将延迟持久性(Lazy Persistency，LP)映射到GPU的方法，并确定了这种映射的设计空间。然后，我们在不同的校验和类型、约简方法、锁定的使用和哈希表设计的情况下，对GPU上的LP性能进行了表征。在深入了解性能瓶颈的基础上，我们提出了一种无哈希表的方法，该方法在成百上千个线程上运行良好，实现了持久化，而对于各种代表性基准测试，性能下降几乎可以忽略不计(2.1%)。我们还提出了一种基于指令的编程语言支持，以简化将LP添加到GPU应用程序的编程工作。
GPUs applications, including many scientific and machine learning applications, increasingly demand larger memory capacity. NVM is promising higher density compared to DRAM and better future scaling potentials. Long running GPU applications can benefit from NVM by exploiting its persistency, allowing crash recovery of data in memory. In this paper, we propose mapping Lazy Persistency (LP) to GPUs and identify the design space of such mapping. We then characterize LP performance on GPUs, varying the checksum type, reduction method, use of locking, and hash table designs. Armed with insights into the performance bottlenecks, we propose a hash table-less method that performs well on hundreds and thousands of threads, achieving persistency with nearly negligible (2.1%) slowdown for a variety of representative benchmarks. We also propose a directive-based programming language support to simplify programming effort for adding LP to GPU applications.