Postprint version. Published in Proceedings of the 3rd Conference on Computing Frontiers: Ischia, Italy, May 3, 2006, pages 83-94.
NOTE: At the time of publication, the author John Oliver was not yet affiliated with Cal Poly.
The definitive version is available at https://doi.org/10.1145/1128022.1128036.
In this paper, we investigate the power implications of tile size selection for tile-based processors. We refer to this investigation as a tile granularity study. This is accomplished by distilling the architectural cost of tiles with different computational widths into a system metric we call the Granularity Indicator (GI). The GI is then compared against the communications exposed when algorithms are partitioned across multiple tiles. Through this comparison, the tile granularity that best fits a given set of algorithms can be determined, reducing the system power for that set of algorithms. When the GI analysis is applied to the Synchroscalar tile architecture, we find that Synchroscalar's already low power consumption can be further reduced by 14% when customized for execution of the 802.11a receiver. In addition, the GI can also be a used to evaluate tile size when considering multiple applications simultaneously, providing a convenient platform for hardware-software co-design.
Electrical and Computer Engineering
This is the author's version of the work. It is posted here by permission of ACM for your personal use. Not for redistribution. The definitive version was published in Proceedings of the 3rd Conference on Computing Frontiers.