OVERVIEW

Research demand for High-performance Computing (HPC) is growing rapidly across more academic disciplines, requiring faster computing, ability to manage and transfer ever larger amounts of data quickly and securely and to a broader community, and shorter front-end time from needs request to computing to science acquisition to knowledge dissemination and technology transfer.

ALIGNMENT

• With the University — National Competitiveness: The quality, capability and performance of HPC resources are critical factors in strengthening the status of UH as a nationally competitive public research university.
• With the University — Local and National Recognition: Advanced computing resources help to secure local and national recognition for UH research achievements.
• With the UH System — Community Advancement: HPC plays a key role in enabling rapid progress from research to technology transfer that directly contributes to Community Advancement.
• With the State — Data Utility: The UIT HPC group’s efforts, together with those of the Hewlett Packard Enterprise Data Science Institute (HPE DSI), directly accomplish the Data Analytics component of TX-DIR’s  Data Utility strategic goal and can promote the Open Data component of the Data Utility goal.

CURRENT STATE

• The UIT-HPC group provides infrastructure support, system administration and backup for computing, as well as storage equipment, including the clusters at HPE DSI, and other equipment (mostly grant funded) housed in the Research Computing Data Center.
• UIT HPC operates as an honest broker for the campus high performance computing community to advise with architecture and design, assist with procurements and implementations and resolve issues when necessary.

PATH TO SUCCESS

• Continue the UIT-HPC/HPE DSI partnership to propel UH research forward by reducing barriers to entry for new researchers interested in High Performance Computing, especially those in less represented areas like arts/humanities, social sciences and business.
• Improve administrative processes to acquire and use research IT resources. Speed is essential not just in computing and handling data, but also in acquiring and implementing technology.

FY2022 INITIATIVES

• Actively advise on emerging directions, such as massively scalable database architecture, cloud-based HPC and Science DMZ for secure, rapid sharing of large datasets over high-speed WAN.
• Improve HPC training experience by developing and providing introductory short courses to support faster on-ramping for researchers new to using HPC resources.
• Promote awareness and use of HPC by under-represented areas, offering guidance in business processes like funding, and not just the technology.
• Support the continued growth of the Cluster Partnership Program with the HPE DSI to provision resources from start-up funds and grants in a highly efficient and cost-effective manner within the existing public clusters at the University.