Hardware-assisted replay of microprocessor programs
Shared-memory parallel programs can be highly non-deterministic due to the unpredictable order in which shared references are satisfied. However, deterministic execution is extremely important for debugging and can also be used for fault-tolerance and other replay-based algorithms. This paper presents a hardware/software design that allows the order of memory references in a parallel program to be logged efficiently by recording a subset of the cache traffic between memory and the CPU's. This log can then be used along with hardware and software control to replay execution. The authors simulate memory and cache traffic for several parallel programs to determine the costs and characteristics of our scheme for realistic applications. The authors then use the information from the simulation to analyze the potential performance under a number of different system parameters.
- OSTI ID:
- 7205533
- Country of Publication:
- United States
- Language:
- English
Similar Records
Adaptive message logging for incremental replay of message-passing programs
Distributed Order Recording Techniques for Efficient Record-and-Replay of Multi-threaded Programs
Related Subjects
990200* -- Mathematics & Computers
ARRAY PROCESSORS
COMPUTER CODES
COMPUTERIZED CONTROL SYSTEMS
COMPUTERIZED SIMULATION
COMPUTERS
CONTROL SYSTEMS
DIGITAL COMPUTERS
FAULT TOLERANT COMPUTERS
MEMORY DEVICES
ON-LINE CONTROL SYSTEMS
ON-LINE SYSTEMS
PARALLEL PROCESSING
PARAMETRIC ANALYSIS
PERFORMANCE
PROGRAMMING
SIMULATION
SYSTEMS ANALYSIS