NASA SBIR 2019-I Solicitation

Proposal Summary

PROPOSAL NUMBER:
 19-1- H6.22-4274
SUBTOPIC TITLE:
 Deep Neural Net and Neuromorphic Processors for In-Space Autonomy and Cognition
PROPOSAL TITLE:
 pMTJ STT-MRAM based Chiplets for Neuromorphic Computing
SMALL BUSINESS CONCERN (Firm Name, Mail Address, City/State/Zip, Phone)
Avalanche Technology
3540 West Warren Avenue
Fremont, CA 94538- 6425
(510) 897-3000

Principal Investigator (Name, E-mail, Mail Address, City/State/Zip, Phone)

Name:
Dr. Rajiv Ranjan
E-mail:
rajiv@avalanche-technology.com
Address:
3450 W Warren Avenue Fremont, CA 94538 - 6451
Phone:
(510) 897-3000

Business Official (Name, E-mail, Mail Address, City/State/Zip, Phone)

Name:
Mr. Danny Sabour
E-mail:
danny@avalanche-technology.com
Address:
3540 W Warren Avenue Fremont, CA 94538 - 6425
Phone:
(510) 897-3000
Estimated Technology Readiness Level (TRL) :
Begin: 1
End: 4
Technical Abstract (Limit 2000 characters, approximately 200 words)

Avalanche Technology will offer MRAM chiplets based on its pMTJ STT-MRAM technology. Avalanche’s MRAM chiplets use Generic MRAM Interface (GMI). The GMI is modeled on the widely-used HBM interface and is optimized for memory accesses. In addition to the memory interface, MRAM chiplets need a second interface to communicate with the rest of the chiplets in a multi-chiplet system. For example, Intel’s Advanced Interface Bus (AIB) protocol is in the forefront for direct chiplet-to-chiplet communication for DARPA’s CHIPS (Common Heterogeneous Integration and Intellectual Property (IP) Reuse Strategies) program. To be used in AIB-based systems, Avalanche’s MRAM chiplet will have an AIB interface and an adapter from the AIB to the HBM interface for the MRAM array.

Potential NASA Applications (Limit 1500 characters, approximately 150 words)

Demonstration of Compute in memory (CIM) for Neuromorphic applications utilizing proven Radiation Hardened STT-MRAM products from Avalanche Technology for Harsh environments.

Potential Non-NASA Applications (Limit 1500 characters, approximately 150 words)

Deep learning and Artificial Intelligence in a wide variety of tasks such as computer vision, image and speech recognition, machine translation, robotics and medical image processing.

Duration: 3
Form Generated on 06/16/2019 23:33:27