DIS: an architecture for fast lisp execution
Thesis/Dissertation
·
OSTI ID:6984117
DIS is an architecture for very fast execution of LISP and other artificial intelligence languages. The DIS architecture uses a number of functional units controlled by a wide (256 bit) instruction. A simulator, compiler,and optimizer were constructed for the DIS architecture. A simulated 100-nanosecond cycle time single-processor DIS machine appears to run LISP on the order of twice as fast as a CRAY-1, and on the order of ten to fifteen times faster than other LISP-directed architectures.
- Research Organization:
- Rensselaer Polytechnic Inst., Troy, NY (USA)
- OSTI ID:
- 6984117
- Resource Relation:
- Other Information: Thesis (Ph. D.)
- Country of Publication:
- United States
- Language:
- English
Similar Records
Architectural approaches to the design of LISP-oriented reduced-instruction set machines
Direct execution of LISP on a list-directed architecture
An efficient and portable SIMD algorithm for charge/current deposition in Particle-In-Cell codes
Thesis/Dissertation
·
Thu Jan 01 00:00:00 EST 1987
·
OSTI ID:6984117
Direct execution of LISP on a list-directed architecture
Conference
·
Mon Mar 01 00:00:00 EST 1982
· Comput. Archit. News; (United States)
·
OSTI ID:6984117
+2 more
An efficient and portable SIMD algorithm for charge/current deposition in Particle-In-Cell codes
Journal Article
·
Mon Sep 19 00:00:00 EDT 2016
· Computer Physics Communications
·
OSTI ID:6984117
+2 more