Samir Lipovaca
ABSTRACT
A method of replacing an explicit loop by the "each" operator was studied. The repetitive, computational part of the loop was extracted into a function FOO. Then, the function was applied to each value of the loop counter vector C (FOO "each" C). Since array operations are optimized on most APL systems, this replacement should produce a most reasonable loop, assuming that in the function FOO array operations were used as much as possible. Two examples of performance measurements showed that this, generally, did not lead to smaller execution time. In one example the "each" operator was faster than an explicit loop while in another the opposite was true.
- Brown, James A., Pakin, Sandra, and Polivka, Raymond P. APL2 at Glance. Prentice Hall, Englewood Cliffs, NJ, 1988. 282. Google Scholar
Digital Library
- Otto, Tilman P. "An APL compiler" APL Berlin 2000 Conference Proceedings APL Quote Quad Vol 30 No 4 Google ScholarDigital Library
- Polivka, Raymond P. and Pakin, Sandra. APL: The Language and Its Usage. Prentice Hall, Englewood Cliffs, NJ, 1975. 397. Google ScholarDigital Library
Index Terms
- Replacing an explicit loop by the "each" operator
Recommendations
Replacing an explicit loop by the "each" operator
A method of replacing an explicit loop by the "each" operator was studied. The repetitive, computational part of the loop was extracted into a function FOO. Then, the function was applied to each value of the loop counter vector C (FOO "each" C). Since ...
Outer-loop vectorization: revisited for short SIMD architectures
PACT '08: Proceedings of the 17th international conference on Parallel architectures and compilation techniquesVectorization has been an important method of using data-level parallelism to accelerate scientific workloads on vector machines such as Cray for the past three decades. In the last decade it has also proven useful for accelerating multi-media and ...
Loop striping: maximize parallelism for nested loops
EUC'06: Proceedings of the 2006 international conference on Embedded and Ubiquitous ComputingThe majority of scientific and Digital Signal Processing (DSP) applications are recursive or iterative. Transformation techniques are generally applied to increase parallelism for these nested loops. Most of the existing loop transformation techniques ...
Comments