The Community for Technology Leaders
Green Image
<p>This paper discusses a new work-scheduling algorithm for parallel search of single-agent state spaces, called <it>Transposition-Table-Driven Work Scheduling</it>, that places the transposition table at the heart of the parallel work scheduling. The scheme results in less synchronization overhead, less processor idle time, and less redundant search effort. Measurements on a 128-processor parallel machine show that the scheme achieves close-to-linear speedups; for large problems the speedups are even superlinear due to better memory usage. On the same machine, the algorithm is 1.6 to 12.9 times faster than traditional work-stealing-based schemes.</p>
Distributed search, single-agent search, work pushing, Transposition-Table-Driven Work Scheduling (TDS), IDA*
John W. Romein, Henri E. Bal, Jonathan Schaeffer, Aske Plaat, "A Performance Analysis of Transposition-Table-Driven Work Scheduling in Distributed Search", IEEE Transactions on Parallel & Distributed Systems, vol. 13, no. , pp. 447-459, May 2002, doi:10.1109/TPDS.2002.1003855
80 ms
(Ver 3.3 (11022016))