Abstract
Architectures for deterministic record-replay (R&R) of multithreaded code are attractive for program debugging, intrusion analysis, and fault-tolerance uses. However, very few of the proposed designs have focused on maximizing replay speed -- a key enabling property of these systems. The few efforts that focus on replay speed require intrusive hardware or software modifications, or target whole-system R&R rather than the more useful application-level R&R.
This paper presents the first hardware-based scheme for unintrusive, application-level R&R that explicitly targets high replay speed. Our scheme, called Cyrus, requires no modification to commodity snoopy cache coherence. It introduces the concept of an on-the-fly software Backend Pass during recording which, as the log is being generated, transforms it for high replay parallelism. This pass also fixes-up the log, and can flexibly trade-off replay parallelism for log size. We analyze the performance of Cyrus using full system (OS plus hardware) simulation. Our results show that Cyrus has negligible recording overhead. In addition, for 8-processor runs of SPLASH-2, Cyrus attains an average replay parallelism of 5, and a replay speed that is, on average, only about 50% lower than the recording speed.
- H. Agrawal et al. An Execution-Backtracking Approach to Debugging. IEEE Software, 8(3), May 1991. Google Scholar
Digital Library
- G. Altekar and I. Stoica. ODR: Output-deterministic replay for multicore debugging. In Symposium on Operating Systems Principles, 2009. Google Scholar
Digital Library
- A. Basu et al. Karma: Scalable deterministic record-replay. In Int. Conference on Supercomputing, 2011. Google Scholar
Digital Library
- B. Bloom. Space/Time Trade-Offs in Hash Coding with Allowable Errors. Communications of the ACM, 11(7), July 1970. Google Scholar
Digital Library
- T. Bressoud and F. Schneider. Hypevisor-based fault-tolerance. ACM Transactions on Computer Systems, 14(1), Feb 1996. Google Scholar
Digital Library
- A. Charlesworth. Starfire: Extending the SMP Envelope. IEEE Micro, 18(1), Jan. 1998. Google Scholar
Digital Library
- Y. Chen et al. LReplay: A pending period based deterministic replay scheme. In Int. Symposium on Computer Architecture, 2010. Google Scholar
Digital Library
- J.-D. Choi and H. Srinivasan. Deterministic replay of Java multithreaded applications. In Symposium on Parallel and Distributed Tools, 1998. Google Scholar
Digital Library
- B. Cully et al. Remus: High availability via asynchronous virtual machine replication. In USENIX Symposium on Networked Systems Design and Implementation, 2008. Google Scholar
Digital Library
- G. W. Dunlap et al. ReVirt: Enabling intrusion analysis through virtual-machine logging and replay. In Symposium on Operating Systems Design and Implementation, 2002. Google Scholar
Digital Library
- G. W. Dunlap et al. Execution replay of multiprocessor virtual machines. In Int. Conference on Virtual Execution Environments, 2008. Google Scholar
Digital Library
- A. Forin. Debugging of heterogeneous parallel systems. In Workshop on Parallel and Distributed Debugging, 1988. Google Scholar
Digital Library
- D. R. Hower and M. D. Hill. Rerun: Exploiting Episodes for Lightweight Memory Race Recording. In Int. Symposium on Computer Architecture, 2008. Google Scholar
Digital Library
- S. T. King and P. M. Chen. Backtracking intrusions. In Symposium on Operating Systems Principles, 2003. Google Scholar
Digital Library
- S. T. King et al. Debugging operating systems with time-traveling virtual machines. In USENIX Annual Technical Conference, 2005. Google Scholar
Digital Library
- O. Laadan et al. Transparent, lightweight application execution replay on commodity multiprocessor operating systems. In SIGMETRICS Int. Conference on Measurement and Modeling of Computer Systems, 2010. Google Scholar
Digital Library
- L. Lamport. Time, clocks, and the ordering of events in a distributed system. Communications of the ACM, 21(7), July 1978. Google Scholar
Digital Library
- T. J. LeBlanc and J. M. Mellor-Crummey. Debugging Parallel Programs with Instant Replay. IEEE Trans. Comput., 36(4), Apr. 1987. Google Scholar
Digital Library
- D. Lee et al. Offline symbolic analysis for multi-processor execution replay. In Int. Symposium on Microarchitecture, 2009. Google Scholar
Digital Library
- D. Lee et al. Respec: Efficient online multiprocessor replayvia speculation and external determinism. In Int. Conference on Architectural Support for Programming Languages and Operating Systems, 2010. Google Scholar
Digital Library
- D. Lee et al. Offline symbolic analysis to infer Total Store Order. In Int. Symposium on High Performance Computer Architecture, 2011. Google Scholar
Digital Library
- P. S. Magnusson et al. Simics: A Full System Simulation Platform. IEEE Computer, 35(2), February 2002. Google Scholar
Digital Library
- P. Montesinos et al. DeLorean: Recording and Deterministically Replaying Shared-Memory Multiprocessor Execution Efficiently. In Int. Symposium on Computer Architecture, 2008. Google Scholar
Digital Library
- P. Montesinos et al. Capo: A software-hardware interface for practical deterministic multiprocessor replay. In Int. Conference on Architectural Support for Programming Languages and Operating Systems, 2009. Google Scholar
Digital Library
- S. Narayanasamy et al. BugNet: Continuously Recording Program Execution for Deterministic Replay Debugging. In Int. Symposium on Computer Architecture, 2005. Google Scholar
Digital Library
- S. Narayanasamy et al. Recording shared memory dependencies using strata. In Int. Conference on Architectural Support for Programming Languages and Operating Systems, 2006. Google Scholar
Digital Library
- D. Z. Pan and M. A. Linton. Supporting reverse execution for parallel programs. In Workshop on Parallel and Distributed Debugging, 1988. Google Scholar
Digital Library
- S. Park et al. PRES: Probabilistic replay with execution sketching on multiprocessors. In Symposium on Operating Systems Principles, 2009. Google Scholar
Digital Library
- H. Patil et al. PinPlay: A framework for deterministic replay and reproducible analysis of parallel programs. In Int. Symposium on Code Generation and Optimization, 2010. Google Scholar
Digital Library
- G. Pokam et al. Architecting a chunk-based memory race recorder in modern CMPs. In Int. Symposium on Microarchitecture, 2009. Google Scholar
Digital Library
- G. Pokam et al. CoreRacer: A practical memory race recorder for multicore x86 TSO processors. In Int. Symposium on Microarchitecture, 2011. Google Scholar
Digital Library
- M. Russinovich and B. Cogswell. Replay for concurrent non-deterministic shared-memory applications. In Programming Language Design and Implementation, 1996. Google Scholar
Digital Library
- S. M. Srinivasan et al. Flashback: a lightweight extension for rollback and deterministic replay for software debugging. In USENIX Annual Technical Conference, 2004. Google Scholar
Digital Library
- K. Veeraraghavan et al. DoublePlay: Parallelizing sequential logging and replay. In Int. Conference on Architectural Support for Programming Languages and Operating Systems, 2011. Google Scholar
Digital Library
- G. Voskuilen et al. Timetraveler: Exploiting acyclic races for optimizing memory race recording. In Int. Symposium on Computer Architecture, 2010. Google Scholar
Digital Library
- M. Xu et al. A "flight data recorder" for enabling full-system multiprocessor deterministic replay. In Int. Symposium on Computer Architecture, 2003. Google Scholar
Digital Library
- M. Xu et al. A regulated transitive reduction (RTR) for longer memory race recording. In Int. Conference on Architectural Support for Programming Languages and Operating Systems, 2006. Google Scholar
Digital Library
- M. Yuffe et al. A fully integrated multi-CPU, GPU and memory controller 32nm processor. In Int. Solid-State Circuits Conference, 2011.Google Scholar
Cross Ref
Index Terms
Cyrus: unintrusive application-level record-replay for replay parallelism
Recommendations
Cyrus: unintrusive application-level record-replay for replay parallelism
ASPLOS '13Architectures for deterministic record-replay (R&R) of multithreaded code are attractive for program debugging, intrusion analysis, and fault-tolerance uses. However, very few of the proposed designs have focused on maximizing replay speed -- a key ...
Cyrus: unintrusive application-level record-replay for replay parallelism
ASPLOS '13: Proceedings of the eighteenth international conference on Architectural support for programming languages and operating systemsArchitectures for deterministic record-replay (R&R) of multithreaded code are attractive for program debugging, intrusion analysis, and fault-tolerance uses. However, very few of the proposed designs have focused on maximizing replay speed -- a key ...
Transparent mutable replay for multicore debugging and patch validation
ASPLOS '13We present Dora, a mutable record-replay system which allows a recorded execution of an application to be replayed with a modified version of the application. This feature, not available in previous record-replay systems, enables powerful new ...







Comments