National systems: Difference between revisions

Jump to navigation Jump to search
replace Types of compute clusters table with text description
No edit summary
(replace Types of compute clusters table with text description)
Line 5: Line 5:


<!--T:3-->
<!--T:3-->
All ''general-purpose'' systems are similar with some differences in interconnect and the number of small memory, large memory and GPU nodes. Niagara is a different system and is designed for ''large parallel'' jobs.
A ''general-purpose'' cluster is designed to support a wide variety of types of jobs, and is composed of a mixture of different nodes.  We broadly classify the nodes as:
* ''base'' nodes, containing typically about 4GB of memory per core;
* ''large-memory'' nodes, containing typically more than 8GB memory per core;
* ''GPU'' nodes, which contain [https://en.wikipedia.org/wiki/Graphics_processing_unit graphic processing units].


<!--T:5-->
The ''large parallel'' cluster [[Niagara]] is designed to support multi-node parallel jobs requiring more than 1000 CPU cores, although jobs as small as a single node are also supported there.  Niagara is composed of nodes of a uniform design, with an interconnect optimized for large jobs.
All systems have large, high-performance attached storage; see the relevant cluster page for more details.


===Types of compute clusters=== <!--T:12-->
All clusters have large, high-performance storage attached.  For details about storage, memory, CPU model and count, GPU model and count, and the number of nodes at each site, please see the individual cluster pages linked under "Resources" at left.
 
<!--T:13-->
{| class="wikitable"
|-
! Type !! Type of nodes !! Type of jobs
|-
| General-purpose
|
* Small memory nodes
* Regular nodes (4GB/core)
* Large memory nodes
* GPU nodes
|
* Serial and small parallel jobs
* Small or large memory jobs
* GPU jobs
|-
| Large parallel
|
* Uniform nodes (4GB/core)
* Optimized interconnect
|
* Parallel jobs requiring +1000 cores
|}


===List of compute clusters=== <!--T:14-->
===List of compute clusters=== <!--T:14-->
Bureaucrats, cc_docs_admin, cc_staff
2,915

edits

Navigation menu