An empirical study of automatic restructuring of nonnumerical programs for parallel processors
IEEE Transactions on Computers
Journal of the ACM (JACM)
Practical Parallel Band Triangular System Solvers
ACM Transactions on Mathematical Software (TOMS)
Structure of Computers and Computations
Structure of Computers and Computations
Control and data dependence for program transformations.
Control and data dependence for program transformations.
Optimization and interconnection complexity for: parallel processors, single-stage networks, and decision trees
Optimizing supercompilers for supercomputers
Optimizing supercompilers for supercomputers
A finite state machine based format model of software pipelined loops with conditions
Progress in computer research
The Journal of Supercomputing
Hi-index | 0.00 |
This paper considers automatic restructuring of loops with conditional branching for parallel processing, especially a class of loops termed 驴conditional cyclic loops.驴 A conditional cyclic loop possesses a dependence cycle caused by conditional branching across loop iterations, which makes it difficult to parallelize. In general, parallel execution of a conditional cyclic loop provides little benefit due to the need of solving a full-order nonlinear Boolean recurrence relation. However, the Boolean recurrence in practice is often of simpler forms. With the simpler forms, the number of possible predicate values of conditional branching is reduced drastically compared to a general conditional cyclic loop. These simple forms of conditional cyclic loops found in practice can be parallelized for $O(p/\log p)$ speedup with $p$ processors.Index Terms驴Automatic program restructuring, conditional cyclic loop, data dependence, linear mixed recurrence loop, parallel algorithm, postfix-IF loop.