Advanced numerical methods for materials engineering and design

Abstract:	We propose to develop a database of solid materials, computationally predicted to be superconductors at different conditions, and use it to train some deep-learning (DL) models. This database will be highly diverse and comprehensive, containing the atomic structures of thousands of materials, and $\lambda$ and $\omega_{\log}$, two parameters of the electron-phonon interactions using which potential superconductivity may be predicted. The diversity of the database will be emphasized, giving some bias toward materials that are connected with high critical temperature. The targeted DL models will be deployed in www.matsml.org and made accessible to the academic research community.

2023	ACCESS Credits	1,500,000.0 ACCESS Credits
2022	SDSC Expanse Projects Storage	1,000.0 GB
2022	SDSC Expanse CPU	1,500,000.0 Core-hours
The estimated value of these awarded resources is $6,650.00. The allocation of these resources represents a considerable investment by the NSF in advanced computing infrastructure for the U.S. The dollar value of the allocation is estimated from the NSF awards supporting the allocated resources.

2020	SDSC Expanse Projects Storage	500.0 GB
2020	SDSC Expanse CPU	650,000.0 Core-hours
2020	SDSC Dell Cluster with Intel Haswell Processors (Comet)	155,000.0 SUs
2020	SDSC Medium-term disk storage (Data Oasis)	500.0 GB
The estimated value of these awarded resources is $5,392.00. The allocation of these resources represents a considerable investment by the NSF in advanced computing infrastructure for the U.S. The dollar value of the allocation is estimated from the NSF awards supporting the allocated resources.
2019	SDSC Dell Cluster with Intel Haswell Processors (Comet)	1,291,749.0 SUs
2019	SDSC Medium-term disk storage (Data Oasis)	500.0 GB
The estimated value of these awarded resources is $19,558.24. The allocation of these resources represents a considerable investment by the NSF in advanced computing infrastructure for the U.S. The dollar value of the allocation is estimated from the NSF awards supporting the allocated resources.
2017	TACC Long-term tape Archival Storage (Ranch)	1,024.0 GB
2017	SDSC Medium-term disk storage (Data Oasis)	1,024.0 GB
2017	SDSC Dell Cluster with Intel Haswell Processors (Comet)	273,802.0 SUs
2017	TACC Dell/Intel Knights Landing, Skylake System (Stampede2)	17,523.0 Node Hours
The estimated value of these awarded resources is $9,079.94. The allocation of these resources represents a considerable investment by the NSF in advanced computing infrastructure for the U.S. The dollar value of the allocation is estimated from the NSF awards supporting the allocated resources.

Development and applications of advanced numerical methods for materials engineering and design