109 urn:lsid:arphahub.com:pub:3dc5f44e-8666-58db-bc76-a455210e8891 JUCS - Journal of Universal Computer Science jucs 0948-695X 0948-6968 Journal of Universal Computer Science 10.3217/jucs-004-02-0099 27465 Research Article The SNAP Project: Building Validated Floating Point Units Al-Twaijry Hesham A. hesham@umunhum.stanford.edu Oberman Stuart F. Fu Steve T. Flynn Michael J. Computer Systems Laboratory, Stanford University, Stanford, Computer Systems Laboratory, Stanford University Stanford

Corresponding author: Hesham A. Al-Twaijry (hesham@umunhum.stanford.edu).

Academic editor:

1998 28 02 1998 4 2 99 109 8C5D9B57-DEE8-505E-8457-A2BF602DE32A 6995506 Hesham A. Al-Twaijry, Stuart F. Oberman, Steve T. Fu, Michael J. Flynn This article is freely available under the J.UCS Open Content License.

SNAP - The Stanford Sub-nanosecond arithmetic processor is an interdisciplinary effort to develop validated theory, and tools for realizing an arithmeti c processor with execution rates under 1ns. The project has targeted the full spectrum of tradeoffs from algorithms, circuit optimizations, system issues, and development of metrics to characterize processors.