Supporting reverse execution for parallel programs

ABSTRACT

Parallel programs are difficult to debug because they run for a, long time and two executions may yield different results. Reverse execution, is a simple and powerful concept that solves both these problems. We are designing a tool for debugging parallel programs, called Recap, that provides the illusion of reverse execution using checkpoints and event recording and playback. During normal execution, Recap logs the results of system calls and shared memory reads: as well as the times that asynchronous events (signals) occur. Recap periodically checkpoints the state of a process by forking and suspending a new process. To reverse execute to a certain point in time, Recap continues the nearest checkpoint process forward in a self-contained environment, simulating all events using the log. We are implementing Recap as part of a larger environment for parallel program development.

REFERENCES

• 1: A. Agarwal , R. L. Sites , M. Horowitz, ATUM: a new technique for capturing address traces using microcode, Proceedings of the 13th annual international symposium on Computer architecture, p.119-127, June 02-05, 1986, Tokyo, Japan
• 2: T. A. Cargill , B. N. Locanthi, Cheap hardware support for software debugging and profiling, Proceedings of the second international conference on Architectual support for programming languages and operating systems, p.82-83, October 1987, Palo Alto, California, United States
• 3: R. Curt, is and L. Wittie, "Bugnet: A Debugging System for Pa.rallel Programming Environments", Proceedings of the 3rd Interna.tional Conference on Distributed Computing Systems, Miami, Florida, October 1982, pp. 394-399.
• 4: Stuart I. Feldman , Channing B. Brown, IGOR: a system for program debugging via reversible execution, Proceedings of the 1988 ACM SIGPLAN and SIGOPS workshop on Parallel and distributed debugging, p.112-123, May 05-06, 1988, Madison, Wisconsin, United States
• 5: T. J. LeBlanc , J. M. Mellor-Crummey, Debugging parallel programs with instant replay, IEEE Transactions on Computers, v.36 n.4, p.471-482, April 1987 [doi>10.1109/TC.1987.1676929]
• 6: M. A. Linton, "Distributed Management of a Software Database", IEEE Software, Vol. 4, No. 6, November 1987, pp 70-76.
• 7: B. P. Miller and Jong-Deok Choi, "A Mechanism for Efficient Debugging of Parallel Programs", Technical l~eport~ TR754, University of Wisconsin-Madison, 1987.
• 8: M. Young , A. Tevanian , R. Rashid , D. Golub , J. Eppinger, The duality of memory and communication in the implementation of a multiprocessor operating system, Proceedings of the eleventh ACM Symposium on Operating systems principles, p.63-76, November 08-11, 1987, Austin, Texas, United States

INDEX TERMS

Primary Classification:
  D. Software
  D.2 SOFTWARE ENGINEERING

Additional Classification:
  C. Computer Systems Organization
  C.1 PROCESSOR ARCHITECTURES
      C.1.2 Multiple Data Stream Architectures (Multiprocessors)
          Subjects: Parallel processors**

  D. Software
  D.4 OPERATING SYSTEMS
      D.4.5 Reliability
          Subjects: Fault-tolerance

General Terms:
Theory