Parallel simulation of parallel programs for large datasets has been shown to offer significant reduction in the execution time of many discrete event models. The paper describes the design and implementation of MPI-SIM, a library for the execution driven parallel simulation of task and data parallel programs. MPI-SIM can be used to predict the performance of existing programs written using MPI for message passing, or written in UC, a data parallel language, compiled to use message passing. The simulation models can be executed sequentially or in parallel. Parallel execution of the models are synchronized using a set of asynchronous conservative protocols. The paper demonstrates how protocol performance is improved by the use of application-level, runtime analysis. The analysis targets the communication patterns of the application. We show the application-level analysis for message passing and data parallel languages. We present the validation and performance results for the simulator for a set of applications that include the NAS Parallel Benchmark suite. The application-level optimization described in the paper yielded significant performance improvements in the simulation of parallel programs, and in some cases completely eliminated the synchronizations in the parallel execution of the simulation model.

Nonblocking algorithms for concurrent objects guarantee that an object is always accessible, in contrast to blocking algorithms in which a slow or halted process can render part or all of the data structure inaccessible to other processes. A number of algorithms have been proposed for shared FIFO queues, but nonblocking implementations are few and either limit the concurrency or provide inefficient solutions. The authors present a simple and efficient nonblocking shared FIFO queue algorithm with O(n) system latency, no additional memory requirements, and enqueuing and dequeuing times independent of the size of the queue. They use the compare & swap operation as the basic synchronization primitive. They model their algorithm analytically and with a simulation, and compare its performance with that of a blocking FIFO queue. They find that the nonblocking queue has better performance if processors are occasionally slow, but worse performance if some processors are always slower than others.>

Nonblocking algorithms for concurrent data structures guarantee that a data structure is always accessible. This is in contrast to blocking algorithms in which a slow or halted process can render part or all of the data structure inaccessible to other processes.