Difference between revisions of "SST-Core"
Jump to navigation
Jump to search
Moskalenko (talk | contribs) |
Moskalenko (talk | contribs) m (Text replacement - "#uppercase" to "uc") |
||
| (One intermediate revision by the same user not shown) | |||
| Line 11: | Line 11: | ||
|{{#vardefine:faq|}} <!--FAQ--> | |{{#vardefine:faq|}} <!--FAQ--> | ||
|{{#vardefine:citation|}} <!--CITATION--> | |{{#vardefine:citation|}} <!--CITATION--> | ||
| − | |{{#vardefine:installation|}} <!--INSTALLATION--> | + | |{{#vardefine:installation|1}} <!--INSTALLATION--> |
|} | |} | ||
<!--BODY--> | <!--BODY--> | ||
| Line 34: | Line 34: | ||
--> | --> | ||
==System Variables== | ==System Variables== | ||
| − | * HPC_{{ | + | * HPC_{{uc:{{#var:app}}}}_DIR - installation directory |
<!--Configuration--> | <!--Configuration--> | ||
{{#if: {{#var: conf}}|==Configuration== | {{#if: {{#var: conf}}|==Configuration== | ||
Latest revision as of 21:24, 6 December 2019
Description
SST is an open source, cross platform simulation platform that provides a framework to connect multiple simulated hardware objects including CPUs, network, memory, etc. Simulations using the toolkit can be run either single node, or run on multiple nodes via MPI. The toolkit provides a parallel discrete event core as well as several programming interfaces including classes to manage random number generation, statistics handling, simulation output and efficient memory pooling for simulation events. The most recent performance evaluation has shown that SST can scale to simulate beyond 1.5M objects and operate efficiently on simulations up to 128 dual-processor nodes.
Required Modules
Parallel (MPI)
- gcc/5.2.0
- openmpi/1.10.2
- sst-core
System Variables
- HPC_SST-CORE_DIR - installation directory
Installation
See the SST-Core_Install page for sst-core installation notes.