Efficient Evaluation of Array Subscripts of Arrays

Hassitt, A.; Lyon, L. E.

doi:10.1147/rd.161.0045

Cited by 12 publications

(2 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While the original APL group was working on the design and development of APLSV in Philadelphia, a rather different line of inquiry was going on in IBM's Palo Alto Scientific Center in California. It introduced the use of arithmetic progression vectors (APv), which conserved both time and storage in many common situations, and facilitated more efficient evaluation of certain array transformations; 20 it made use of a very fast syntax analyzer that required a new internal representation of APL statements; and it used a different storage allocation method. As shown in the first column of Figure 1, this work first resulted in a microcoded APL system for the System/360 Model 25.…”

Section: Falkoffmentioning

confidence: 99%

The IBM family of APL systems

Falkoff

1991

IBM Syst. J.

View full text Add to dashboard Cite

The developmental history of IBM subfamilies of APL systems is traced in this paper, focusing on the inter-relationships among them and the. methods of implementation used by the venous groups Involved. The language Itself, and.the way Its evolution was managed, are also considered as factors Influencing the development process. A chart Is Included that illustrates the evolution of mainframe and small machine programming products supporting APL, beginning in 1964 up to the present time.

show abstract

Section: Falkoffmentioning

confidence: 99%

The IBM family of APL systems

Falkoff

1991

IBM Syst. J.

View full text Add to dashboard Cite

show abstract

“…In addition, APL never had an associated indexing calculus like the ψ-calculus. Similarly, closure was not obtained for Abrams' indexing rules despite 10+ years of research [73,74,75,76,77].…”

Section: Introductionmentioning

confidence: 98%

Conformal Computing: Algebraically connecting the hardware/software boundary using a uniform approach to high-performance computation for software and hardware applications

Mullin,

Raynolds

2008

Preprint

View full text Add to dashboard Cite

Springer General Radix ExperimentsThe SPARCserver 1000E and the SUN 20 were both machines dedicated to running our experiments. We ran our experiments for a variety of different radices, given the following constraints on the input vector length:In determining what results to discuss and represent in our graphs, we removed the following radices: 128, and 256. Their non-competitive performance Conclusions for the General Radix ApproachWe have succeeded in simplifying a solution to a complex problem: faster and bigger FFTs. Reproducible experiments indicate that our designs outperform all tested packages in either all, or a majority of the cases, while remaining competitive in the others.

show abstract