Transactional Memory for Reliability

作者: Gulay Yalcin , Osman Unsal

DOI: 10.1007/978-3-319-14720-8_13

关键词: Embedded systemKey (cryptography)UndoSoft errorDatabase transactionError detection and correctionTransient (computer programming)Transactional memoryComputer scienceReliability (computer networking)

摘要: It is foreseen that technology trends will increase the transient and permanent fault rates in future processors. Thus providing reliability for both applications running on personal computers mission-critical systems becoming an absolute necessity. A reliable system requires inclusion of two key capabilities: 1) error detection 2) recovery mechanisms. Transactional Memory (TM) provides ideal base recovery. First, TM mechanisms to abort transactions case a conflict, thus they discard or undo all tentative memory updates restart execution from beginning transaction. Thus, transaction’s start can be viewed as locally checkpointed stable state which used Second, transactional semantics allows deferred until transaction commits (or value becomes externally visible), so cost reduced compared traditional schemes (in conducted et every instruction [26]) while its efficiency increased.

参考文章(36)
Shubu Mukherjee, Architecture Design for Soft Errors ,(2008)
Pascal Felber, Christof Fetzer, Torvald Riegel, Composable Error Recovery With Transactional Memory. Bulletin of The European Association for Theoretical Computer Science. ,vol. 99, pp. 105- 133 ,(2009)
Pascal Felber, Jons-Tobias Wamhoff, Christof Fetzer, Transactional Memory for Dependable Embedded Systems (Poster) hot topics in system dependability. ,(2011)
Matthias A. Blumrich, Martin Ohmacht, Burkhard Steinmacher-Burow, Krishnan Sugavanam, Alan Gara, Dong Chen, Mark E. Giampapa, Philip Heidelberger, Local rollback for fault-tolerance in parallel computing systems ,(2010)
Jons-Tobias Wamhoff, Mario Schwalbe, Rasha Faqeh, Christof Fetzer, Pascal Felber, Transactional Encoding for Tolerating Transient Hardware Errors international symposium on stabilization safety and security of distributed systems. pp. 1- 16 ,(2013) , 10.1007/978-3-319-03089-0_1
Richard M. Yoo, Christopher J. Hughes, Konrad Lai, Ravi Rajwar, Performance evaluation of Intel® transactional synchronization extensions for high-performance computing ieee international conference on high performance computing data and analytics. pp. 19- ,(2013) , 10.1145/2503210.2503232
Lance Hammond, Vicky Wong, Mike Chen, Brian D. Carlstrom, John D. Davis, Ben Hertzberg, Manohar K. Prabhu, Honggo Wijaya, Christos Kozyrakis, Kunle Olukotun, Transactional Memory Coherence and Consistency ACM SIGARCH Computer Architecture News. ,vol. 32, pp. 102- 113 ,(2004) , 10.1145/1028176.1006711
Nematollah Bidokhti, SEU concept to reality (allocation, prediction, mitigation) reliability and maintainability symposium. pp. 1- 5 ,(2010) , 10.1109/RAMS.2010.5448078
Layali Rashid, Karthik Pattabiraman, Sathish Gopalakrishnan, Towards understanding the effects of intermittent hardware faults on programs dependable systems and networks. pp. 101- 106 ,(2010) , 10.1109/DSNW.2010.5542613
Robert L. Bocchino, Vikram S. Adve, Bradford L. Chamberlain, Software transactional memory for large scale clusters Proceedings of the 13th ACM SIGPLAN Symposium on Principles and practice of parallel programming - PPoPP '08. pp. 247- 258 ,(2008) , 10.1145/1345206.1345242