An automatic scratch pad memory management tool and MPEG-4 encoder case study

Authors:
Rogier Baert;Eddy de Greef;Erik Brockmeyer
Affiliations:
IMEC, Kapeldreef, Leuven, Belgium;IMEC, Kapeldreef, Leuven, Belgium;IMEC, Kapeldreef, Leuven, Belgium
Venue:
Proceedings of the 45th annual Design Automation Conference
Year:
2008

Citing 5
Cited 5

Data Reuse Analysis Technique for Software-Controlled Memory Hierarchies

Proceedings of the conference on Design, automation and test in Europe - Volume 1
Layer Assignment echniques for Low Energy in Multi-Layered Memory Organisations

DATE '03 Proceedings of the conference on Design, Automation and Test in Europe - Volume 1
Architecture Exploration for a Reconfigurable Architecture Template

IEEE Design & Test
An integrated scratch-pad allocator for affine and non-affine code

Proceedings of the conference on Design, automation and test in Europe: Proceedings
Dynamic allocation for scratch-pad memory using compile-time decisions

ACM Transactions on Embedded Computing Systems (TECS)

Design and Tool Flow of Multimedia MPSoC Platforms

Journal of Signal Processing Systems
Challenging VMs on battery-powered embedded devices

Proceedings of the Third Workshop on Virtual Machines and Intermediate Languages
A framework for automatic parallelization, static and dynamic memory optimization in MPSoC platforms

Proceedings of the 47th Design Automation Conference
An Efficient Memory Organization for High-ILP Inner Modem Baseband SDR Processors

Journal of Signal Processing Systems
Combined loop transformation and hierarchy allocation for data reuse optimization

Proceedings of the International Conference on Computer-Aided Design

Quantified Score

Hi-index	0.00

Visualization

Abstract

Using software-controlled Scratch-Pad Memory (SPM) in Systems-on-Chip has the potential of reducing power consumption by using design-time application knowledge to reduce memory accesses and processor stalls. This paper presents a fully automatic application analysis and transformation tool which selects data-structures for transfer to the SPM and schedules data transfers between background memory and SPM (pre-fetching) to achieve both high performance and low power consumption. A case study applying this tool on an MPEG-4 video encoder shows an overall power reduction of 25%, a 40% power reduction in just the memories and a 40% reduction in processor cycles as compared to an optimized hardware-cache based solution.