Cooperative application/OS DRAM fault recovery.

Ferreira, Kurt Brian ; Bridges, Patrick G. (University of New Mexico, Albuquerque, NM) ; Heroux, Michael Allen ; Hoemmen, Mark ; Brightwell, Ronald Brian

Sandia National Laboratories

关键词: Computers;    Convergence;    Programming;    Computer Codes;    99 General And Miscellaneous//Mathematics, Computing, And Information Science;   

DOI  :  10.2172/1044954
RP-ID  :  SAND2012-4059
RP-ID  :  AC04-94AL85000
RP-ID  :  1044954

美国|英语

来源: UNT Digital Library

【 摘 要 】

Exascale systems will present considerable fault-tolerance challenges to applications and system software. These systems are expected to suffer several hard and soft errors per day. Unfortunately, many fault-tolerance methods in use, such as rollback recovery, are unsuitable for many expected errors, for example DRAM failures. As a result, applications will need to address these resilience challenges to more effectively utilize future systems. In this paper, we describe work on a cross-layer application/OS framework to handle uncorrected memory errors. We illustrate the use of this framework through its integration with a new fault-tolerant iterative solver within the Trilinos library, and present initial convergence results.