Title :
Optimal 2D Data Partitioning for DMA Transfers on MPSoCs
Author :
Saidi, Selma ; Tendulkar, Pranav ; Lepley, Thierry ; Maler, Oded
Author_Institution :
VERIMAG Lab., Univ. of Grenoble, Grenoble, France
Abstract :
Reducing the effects of off-chip memory access latency is a key factor in exploiting efficiently embedded multicore platforms. We consider architectures that admit a multi-core computation fabric, having its own fast and small memory to which the data blocks to be processed are fetched from external memory using a DMA (direct memory access) engine, employing a double- or multiple-buffering scheme to avoid processor idling. In this paper we focus on application programs that process two dimensional data arrays and we determine automatically the size and shape of the portions of the data array which are subject to a single DMA call, based on hardware and applications parameters. When the computation on different array elements are completely independent, the asymmetry of memory structure leads always to prefer one-dimensional horizontal pieces of memory, while when the computation of a data element shares some data with its neighbors, there is a pressure for more "square" shapes to reduce the amount of redundant data transfers. We provide an analytic model for this optimization problem and validate our results by running a mean filter application on the CELL simulator.
Keywords :
file organisation; multiprocessing systems; system-on-chip; CELL simulator; DMA call; DMA engine; DMA transfers; MPSoC; array elements; data blocks; data element; direct memory access engine; double-buffering scheme; embedded multicore platforms; mean filter application; memory structure asymmetry; multicore computation fabric; multiple-buffering scheme; off-chip memory access latency; one-dimensional horizontal memory pieces; optimal 2D data partitioning; optimization problem; processor idling; two-dimensional data arrays; Arrays; Computational modeling; Memory management; Microprocessors; Shape; CELL Processor; Data paralleization; Direct Memory Access (DMA); Double Buffering;
Conference_Titel :
Digital System Design (DSD), 2012 15th Euromicro Conference on
Conference_Location :
Izmir
Print_ISBN :
978-1-4673-2498-4
DOI :
10.1109/DSD.2012.99