Detecting Silent Data Corruption for Extreme-Scale MPI Applications
Detecting Silent Data Corruption for Extreme-Scale MPI Applications
复制标题
DOI：
10.1145/2802658.2802665
复制
发表时间：
2015-09
期刊：
影响因子：
--
通讯作者：
L. Bautista-Gomez;F. Cappello
中科院分区：
文献类型：
--
作者：
L. Bautista-Gomez;F. Cappello
下一代超级计算机预计将拥有更多的组件，同时每次操作消耗的能源将减少数倍。这些趋势正在将超级计算机构造推向小型化和节能策略的极限。因此，预计软错误的数量在未来几年将急剧增加。虽然已经有了纠正或至少检测一些软错误的机制，但这些错误中有很大一部分没有被硬件注意到。这种无声的错误是非常有害的，因为它们可以使应用程序默默地产生错误的结果。在这项工作中，我们提出了一种技术，利用高性能计算应用程序的某些属性，以检测在应用程序级别的无声错误。我们的技术仅基于应用程序数据集的行为来检测腐败，并且与应用程序无关。我们提出了多个腐败检测器，我们夫妇一起工作的方式透明的用户。我们证明，这种策略可以检测到超过80%的腐败，而产生不到1%的开销。我们表明，假阳性率小于1%，当考虑到多位腐败，检测召回率增加到95%以上。
Next-generation supercomputers are expected to have more components and, at the same time, consume several times less energy per operation. These trends are pushing supercomputer construction to the limits of miniaturization and energy-saving strategies. Consequently, the number of soft errors is expected to increase dramatically in the coming years. While mechanisms are in place to correct or at least detect some soft errors, a significant percentage of those errors pass unnoticed by the hardware. Such silent errors are extremely damaging because they can make applications silently produce wrong results. In this work we propose a technique that leverages certain properties of high-performance computing applications in order to detect silent errors at the application level. Our technique detects corruption based solely on the behavior of the application datasets and is application-agnostic. We propose multiple corruption detectors, and we couple them to work together in a fashion transparent to the user. We demonstrate that this strategy can detect over 80% of corruptions, while incurring less than 1% of overhead. We show that the false positive rate is less than 1% and that when multi-bit corruptions are taken into account, the detection recall increases to over 95%.