A brief introduction to HPC and research computing at Pomona College.
What is HPC?
- HPC means High Performance Computing, Highly Parallel Computing, or “supercomputing.”
- HPC systems solve problems faster than standard desktop computers, especially when problems have large datasets, or require many cycles of calculations.
- HPC systems are clustered, or interconnected, multiple high-end computers known as servers.
- An HPC system is typically shared among many users.
How Does HPC Work?
- Each server in an HPC cluster is called a “node.”
- Each node has multiple “processing units” with multiple “cores.”
- Each core is capable of making independent calculations which speeds up calculations.
- Users interact with the HPC system from their own computers.
- A head node and a scheduler manage the shared use of the HPC system.
What Are HPC Use Cases?
- Biology – genomics and transcriptomics
- Neuroscience – brain scan analyses
- Physics – galaxies colliding calculations
- Economics – prediction models
- Linguistics – large text database analyses
- Art – large image database analyses
- Chemistry – 3D molecular structure visualization
- Environmental analysis – climate change models
How does HPC fit in Pomona College’s research support efforts?
Faculty research is a complex activity involving proposals, labs and technology, staff and support, collaborators, data collection and management, publishing.
Academic Affairs supports faculty research through policies, grants, funding, SURP, IRB, tenure, budgets.
ITS supports research with technology including infrastructure, hardware, software, data storage and backup, security, account management, cloud services, technical support.
Not all research projects will need the power of HPC; however, it will become an important tool as many research areas become more complex.