RaPiD { A Con gurable Computing Architecture for Compute - Intensive Applications

Con gurable computers have attracted considerable attention recently because they promise to deliver the performance of application-speci c hardware along with the exibility of general-purpose computers. Unfortunately, con gurable computing has had rather limited success to date. We believe that the FPGAs currently used to construct con gurable computers are too general to achieve good cost-performance on computationally-intensive applications that demand special-purpose hardware. This paper describes a new architecture called RaPiD (Recon gurable Pipelined Datapaths), which is optimized for highly repetitive, computationally-intensive tasks. Very deep application-speci c computation pipelines can be con gured in RaPiD that deliver very high performance for a wide range of applications. RaPiD achieves this using a coarse-grained recon gurable architecture that mixes the appropriate amount of static con guration with dynamic control.