Fault tolerant matrix-matrix multiplication: correcting soft errors on-line
Fault tolerant matrix-matrix multiplication: correcting soft errors on-line
复制标题
容错矩阵-矩阵乘法：在线纠正软错误
DOI：
--
复制
发表时间：
2011
期刊：
影响因子：
--
通讯作者：
Zizhong Chen
中科院分区：
文献类型：
--
作者：
Panruo Wu;Chong Ding;Longxiang Chen;Feng Gao;T. Davies;Christer Karlsson;Zizhong Chen
软错误是一次性事件，会破坏计算系统的状态，但不会破坏其整体功能。软错误通常不会中断受影响程序的执行，但受影响的计算结果不再可信。基于算法的容错（ABFT）是一种众所周知的修正矩阵-矩阵乘法软误差的技术。虽然ABFT比三模冗余（TMR）——一种传统的纠正软错误的通用技术——具有更高的效率，但ABFT和TMR都是在计算完成后离线检测错误。本文将传统的ABFT技术从离线扩展到在线，在程序执行过程中可以在计算过程中检测到矩阵-矩阵乘法中的软错误，并通过及时纠正错误计算来提高效率。实验结果表明，与ATLAS dgemm（）相比，所提出的技术可以每十秒纠正一个错误，而性能损失可以忽略不计（即小于1%）。
Soft errors are one-time events that corrupt the state of a computing system but not its overall functionality. Soft errors normally do not interrupt the execution of the affected program, but the affected computation results can not be trusted any more. A well known technique to correct soft errors in matrix-matrix multiplication is algorithm-based fault tolerance (ABFT). While ABFT achieves much better efficiency than triple modular redundancy (TMR) - a traditional general technique to correct soft errors, both ABFT and TMR detect errors off-line after the computation is finished. This paper extends the traditional ABFT technique from off-line to on-line so that soft errors in matrix-matrix multiplication can be detect in the middle of the computation during the program execution and higher efficiency can be achieved by correcting the corrupted computations in a timely manner. Experimental results demonstrate that the proposed technique can correct one error every ten seconds with negligible (i.e., less than 1%) performance penalty over the ATLAS dgemm().