SBIR-STTR Award

Integrated System For Numeric And Symbolic Computation
Award last edited on: 4/10/2014

Sponsored Program
SBIR
Awarding Agency
DOD : Navy
Total Award Amount
$556,332
Award Phase
2
Solicitation Topic Code
N87-159
Principal Investigator
Lee Higbie

Company Information

ANTs Software Inc (AKA: Sullivan Computer Corporation~CHoPP Computer Corporation~ANTs software.com)

801 Mahler Road Suite G
Burlingame, CA 94010
   (650) 692-0219
   info@ants.com
   www.antssoftware.com
Location: Single
Congr. District: 14
County: San Mateo

Phase I

Contract Number: N60921-90-C-0049
Start Date: 00/00/00    Completed: 00/00/00
Phase I year
1990
Phase I Amount
$58,256
The proposed tasks are to investigate the computer features necessary to provide high performance combined numeric and symbolic processing. This effort uses the Chopp supercomputer architecture as the target machine. Chopp is a recent breakthrough in the organization of general purpose computers for numerical performance at the highest level. A single node of the conventional technology Chopp 1 is a uniprocessor that has four times the throughput of the Cray x-mp for sequential Livermore loops benchmarks. This demonstrates that the architecture is an innovative breakthrough. Chopp computer corporation develops and sells four to sixteen processor, air-cooled supercomputers based upon the innovative Chopp architecture. Evaluation of Chopp hardware and language for symbolic processing determines the expected performance, value and effort necessary to produce a combined symbolic and numeric Chopp computer.

Phase II

Contract Number: N60921-90-C-0049
Start Date: 00/00/00    Completed: 00/00/00
Phase II year
1989
(last award dollars: 1990)
Phase II Amount
$498,076

The phase i study showed that a new single parallel computer of unique design, the chopp daspa, can provide outstanding performance for all three application domains of importance in military missions - namely, numerical analysis, symbolic processing and real time computing. These three application domains are currently supported by a heterogeneous collection of general purpose and special purpose computers which collectively do not yet meet the projected performance levels required for future systems. Thus, we have shown that a cluster of identical chopp daspa computers solves many of the problems of the next generation computer resource program without sacrificing any of its benefits. The phase ii effort will accomplish detailed design of the computer system, detailed verification of its performance for typical mission models, and technology transfer to navy system laboratories and interested military vendors.