In this paper, the reproduction of trigonometric polynomials with two-overlapping local cosine bases is investigated. This study is motivated by the need to represent most effectively a Fourier series in the form of a localized cosine series for the purpose of local analysis, thus providing a vehicl
A Matrix-Based Approach to Global Locality Optimization
β Scribed by Mahmut Kandemir; Alok Choudhary; J. Ramanujam; Prith Banerjee
- Publisher
- Elsevier Science
- Year
- 1999
- Tongue
- English
- Weight
- 824 KB
- Volume
- 58
- Category
- Article
- ISSN
- 0743-7315
No coin nor oath required. For personal study only.
β¦ Synopsis
Global locality optimization is a technique for improving the cache performance of a sequence of loop nests through a combination of loop and data layout transformations. Pure loop transformations are restricted by data dependencies and may not be very successful in optimizing imperfectly nested loops and explicitly parallelized programs. Although pure data transformations are not constrained by data dependencies, the impact of a data transformation on an array might be program-wide; that is, it can affect all the references to that array in all the loop nests. Therefore, in this paper we argue for an integrated approach that employs both loop and data transformations. The method enjoys the advantages of most of the previous techniques for enhancing locality and is efficient. In our approach, the loop nests in a program are processed one by one and the data layout constraints obtained
π SIMILAR VOLUMES
A study of mode localization in mistuned bladed disks is performed using transfer matrices. The transfer matrix approach yields the free response of a general, mono-coupled, perfectly cyclic assembly in closed form. A mistuned structure is represented by random transfer matrices, and the expansion o
A new algorithm is presented for the location of the global minimum of a multiple minima problem. It begins with a series of randomly placed probes in phase space, and then uses an iterative Gaussian redistribution of the worst probes into better regions of phase space until all probes converge to a