Architecture Exploration for Efficient Data Transfer and Storage in Data-Parallel Applications - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2010

Architecture Exploration for Efficient Data Transfer and Storage in Data-Parallel Applications

Abstract

Due to the complexity of modern data parallel applications such as image processing applications, automatic approach to infer suitable and efficient hardware realizations are more and more required. Typically, the optimization of data transfer and storage micro-architecture has a key role for the data parallelism. In this paper, we propose a comprehensive method to explore the mapping of a high-level representation of an application into a customizable hardware accelerator. The high-level representation is in a language called Array-OL. The customizable architecture uses FIFO queues and double buffering mechanism to mask the latency of data transfers and external memory access. The mapping of a high-level representation onto the given architecture is performed by applying a set of loop transformations in Array-OL. A method based on integer partition is used to reduce the space of explored solutions.
Fichier principal
Vignette du fichier
main.pdf (452.13 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

inria-00522786 , version 1 (01-10-2010)

Identifiers

Cite

Rosilde Corvino, Abdoulaye Gamatié, Pierre Boulet. Architecture Exploration for Efficient Data Transfer and Storage in Data-Parallel Applications. Euro-Par - 16th International Euro-Par Conference - 2010, Aug 2010, Ischia, Italy. pp.101--116, ⟨10.1007/978-3-642-15277-1_11⟩. ⟨inria-00522786⟩
238 View
150 Download

Altmetric

Share

Gmail Facebook X LinkedIn More