Exploring the VLSI scalability of stream processors
- 27 August 2003
- proceedings article
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 3 6mb, 153-164
- https://doi.org/10.1109/hpca.2003.1183534
Abstract
Stream processors are high-performance programmable processors optimized to run media applications. Recent work has shown these processors to be more area- and energy-efficient than conventional programmable architectures. This paper explores the scalability of stream architectures to future VLSI technologies where over a thousand floating-point units on a single chip will be feasible. Two techniques for increasing the number of ALUs in a streamprocessor are presented: intracluster and intercluster scaling. These scaling techniques are shown to be cost-efficient to tens of ALUs per cluster and to hundreds of arithmetic clusters. A 640-ALU stream processor with 128 clusters and 5 ALUs per cluster is shown to be feasible in 45 nanometer technology, sustaining over 300 GOPS on kernels and providing 15.3x of kernel speedup and 8.0x of application speedup over a 40-ALU stream processor with a 2% degradation in area per ALU and a 7% degradation in energy dissipated per ALU operation.Keywords
This publication has 11 references indexed in Scilit:
- Media processing applications on the Imagine stream processorPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2003
- Vector vs. superscalar and VLIW architectures for embedded multimedia benchmarksPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2003
- Implementation of a third-generation 1.1GHz 64b microprocessorPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2003
- A 0.18 μm CMOS IA32 microprocessor with a 4 GHz integer execution unitPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- The future of wiresProceedings of the IEEE, 2001
- Imagine: media processing with streamsIEEE Micro, 2001
- Efficient conditional operations for data-parallel architecturesPublished by Association for Computing Machinery (ACM) ,2000
- Baring it all to software: Raw machinesComputer, 1997
- Spert-II: a vector microprocessor systemComputer, 1996
- A stereo machine for video-rate dense depth mapping and its new applicationsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1996