Introduction - Why HPC?
|
High Performance Computing (HPC) typically involves connecting to very large computing systems elsewhere in the world.
These other systems can be used to do work that would either be impossible or much slower or smaller systems.
The standard method of interacting with such systems is via a command line interface.
|
Access the HPC
|
Interaction with a HPC system starts by first connecting to a login node.
The login node is not the HPC, just an interactive gateway.
Command-line (terminal) interaction may be established through Secure Shell (SSH).
GUI interaction may be established through VNC.
|
Components of HPC
|
The HPC is made up many computers, referred to as nodes.
Nodes have various CPU and memory resources.
Nodes are grouped together into partitions.
Each partition has a set maximum time-limit allowed for jobs on its nodes.
A job sheduling system administers the entire system.
The scheduler assigns jobs to nodes, and maintains a queue of waiting jobs.
The login node is largely used to interact with the scheduler.
sinfo can list partitions and nodes available.
|
Submit a batch Job
|
|
Resource Management
|
|
Modules
|
|
Data transfer
|
|
File Systems
|
|
Submitting array jobs
|
|