2D-VLIW: An Architecture Based on the Geometry of Computation

  • Authors:

  • Ricardo Santos;Rodolfo Azevedo;Guido Araujo

  • Affiliations:

  • Dom Bosco Catholic University, Campo Grande, MS, Brazil;State University of Campinas Institute of Computing, Brazil;State University of Campinas Institute of Computing, Brazil

  • Venue:
  • ASAP '06 Proceedings of the IEEE 17th International Conference on Application-specific Systems, Architectures and Processors
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

This work proposes a new architecture and execution model called 2D-VLIW. This architecture adopts an execution model based on large pieces of computation running over a matrix of functional units connected by a set of local register spread across the matrix. Experiments using the Mediabench and SPECint00 programs and the Trimaran compiler show performance gains ranging from 5% to 63%, when comparing our proposal to an EPIC architecture with the same number of registers and functional units. We also show that the g721-enc program running on a 2D-VLIW 3脳3 matrix had a speedup of 1.37 over a 2脳2 matrix while the same program over the EPIC processor with 9 functional units had a speedup of 1.12 over an EPIC processor with 4 functional units. For some internal procedures from Mediabench and SPECint programs, the average 2D-VLIW OPC (operations per cycle) was up to 10 times greater than for the equivalent EPIC processor.