Compute Cluster
Technical Benefits and Features
Technical Benefits and Features
Welcome to the HBSGrid! The HBS compute cluster offers more memory, speed, and processing power through a single entry point. Examples of use cases that will enable or accelerate research includes:
- Running software interactively that requires computing power (RAM or CPUs) equal to or more than one's desktop or laptop.
- Running an analysis that may take a few hours, with the ability of freeing up one's desktop or laptop for other work.
- Analyzing or transforming large datasets sitting on the research storage spaces with computing resources local to that data.
- Working with data in the research MariaDB database.
- Storing, analyzing, and visualizing Level 3 or Level 4 sensitive data.
- Running a piece of code hundreds of times for a parameter sweep, optimization, or model fitting.
- You wish to work on research project with collaborators outside of HBS.
Additionally, technical benefits and features of the platform include:
- The ability to optimize resources through load sharing, scheduling, and queuing of jobs across multiple servers
- Access to a browser-based interface for submitting jobs in addition to the usual connection through SSH
- Multiple servers that provide redundancy to avoid a single point of failure
- Up to 10x speedup for certain math, algebraic, and statistical functions & algorithms.
- No restriction on RAM (other than physical machine constraints), and users allowed up to 24 cores at one time, spread across a max of 3 sessions
- No restriction on RAM (other than physical machine constraints) or number of jobs4 , with max of 16 cores for “short” batch jobs, 8 cores for “long” batch jobs, and 8 cores for “infinite” batch jobs3