Reconfigurable Functional Units for Scientific Superscalar Processors
Evans, J.
Rupnow, K.
Compton, K.
Univ. of Wisconsin-Madison, Madison;
This paper appears in: Field-Programmable Technology, 2007. ICFPT 2007. International Conference on
Publication Date: 12-14 Dec. 2007
On page(s): 73-80
Location: Kitakyushu,
ISBN: 978-1-4244-1472-7
INSPEC Accession Number: 9808640
Digital Object Identifier: 10.1109/FPT.2007.4439234
Current Version Published: 2008-01-22
Abstract
As it becomes more difficult to increase single-threaded performance, focus on multi-core processor designs increases. However, individual core performance is still important, especially for long-executing applications such as in scientific computing. Based on scientific application needs as modeled by SPEC-FP and a set of applications from Sandia National Labs, we have created several different reconfigurable functional unit (RFU) designs for superscalar multi-processor supercomputers. This paper discusses the design process and evaluates the RFUs' ability to implement instruction dataflow graphs from scientific workloads. Our best-performing RFU design is able to implement 89% of the dataflow graphs in the benchmarks.
Index
Terms
Available to subscribers and IEEE members.
References
Available to subscribers and IEEE members.
Citing Documents
Available to subscribers and IEEE members.
Cart
