You are here

Chemistream: Big Data Materials on HPC Clouds

Award Information
Agency: Department of Energy
Branch: N/A
Contract: DE-SC0020535
Agency Tracking Number: 249361
Amount: $206,481.00
Phase: Phase I
Program: SBIR
Solicitation Topic Code: 01a
Solicitation Number: DE-FOA-0002145
Solicitation Year: 2020
Award Year: 2020
Award Start Date (Proposal Award Date): 2020-01-06
Award End Date (Contract End Date): 2020-11-17
Small Business Information
5621 Arapahoe Avenue, Suite A, Boulder, CO, 80303-1379
DUNS: 806486692
HUBZone Owned: N
Woman Owned: N
Socially and Economically Disadvantaged: N
Principal Investigator
 Scott Sides
 (303) 718-7028
Business Contact
 Laurence Nelson
Phone: (720) 974-1856
Research Institution
High performance computing HPC) plays a key role in materials science, climate research, energy technology and others. Artificial intelligence AI) and machine learning ML) are currently used for pattern recognition, email filtering, predictive analytics and others. However, recent surveys have shown an under-representation by companies that could leverage HPC technologies. One reason is that simulation codes of interest can be complex to build for local compute clusters. A second reason is that moving away from local workstations to take advantage of HPC resources can be filled with difficulties particularly for small and medium commercial users. We will develop an easy-to-use product ’Chemistream’ that streamlines access to a variety of material properties databases some funded by DOE) to enable artificial intelligence AI) and machine learning ML) to take advantage of HPC cloud computing resources. During Phase I we will investigate how to streamline the use of S3 persistent storage at Amazon Web Services. This storage will be used to cache relevant parts of the desired materials databases used in the other Phase I tasks. In Phase I we will investigate how to use remote cloud resources to extend the current analysis of NREL’s Bond Dissociation Energy BDE) database which uses ML to identify similar bonds in the large database to predict the BDE of new small molecules. We will also investigate how to use AI/ML to predict the partial charges on small molecules using NREL’s organic photovoltaic OPV) database. Lastly, these examples will be integrated into Chemistream, a Tech-X product that currently provides a user-friendly application to manage HPC cloud computing resources. Simulations of advanced materials and energy technologies speed development of novel products and the identification of new domestic energy sources and will help US industries compete and be more efficient. Better utilization of cloud computing resources will promote development of better computing infrastructure generally and broaden the use of DOE funded codes and databases. Support letters from nanotech company ForgeNano Louisville, CO) and solar energy company NextEnergy Santa Barbara, CA) are indicators of the broad appeal of AI/ML capabilities on clouds addressed by this Phase I.

* Information listed above is at the time of submission. *

US Flag An Official Website of the United States Government