Skip to content
GitLab
Explore
Sign in
Primary navigation
Search or go to…
Project
hpc-compendium
Manage
Activity
Members
Labels
Plan
Issues
Issue boards
Milestones
Wiki
Code
Merge requests
Repository
Branches
Commits
Tags
Repository graph
Compare revisions
Snippets
Deploy
Releases
Package Registry
Container Registry
Model registry
Operate
Terraform modules
Monitor
Incidents
Service Desk
Analyze
Value stream analytics
Contributor analytics
Repository analytics
Model experiments
Help
Help
Support
GitLab documentation
Compare GitLab plans
Community forum
Contribute to GitLab
Provide feedback
Terms and privacy
Keyboard shortcuts
?
Snippets
Groups
Projects
Show more breadcrumbs
ZIH
hpcsupport
hpc-compendium
Commits
17f59897
Commit
17f59897
authored
2 years ago
by
Martin Schroschk
Committed by
Noah Löwer
2 years ago
Browse files
Options
Downloads
Patches
Plain Diff
Apply 3 suggestion(s) to 1 file(s)
parent
f021d0e7
No related branches found
No related tags found
2 merge requests
!808
Automated merge from preview to main
,
!767
Add new file spec.md to software - menu item Performance Engineering Tools
Changes
1
Hide whitespace changes
Inline
Side-by-side
Showing
1 changed file
doc.zih.tu-dresden.de/docs/software/spec.md
+5
-8
5 additions, 8 deletions
doc.zih.tu-dresden.de/docs/software/spec.md
with
5 additions
and
8 deletions
doc.zih.tu-dresden.de/docs/software/spec.md
+
5
−
8
View file @
17f59897
...
@@ -8,8 +8,8 @@ system 'Taurus' (partition `haswell`) is the benchmark's reference system denoti
...
@@ -8,8 +8,8 @@ system 'Taurus' (partition `haswell`) is the benchmark's reference system denoti
The tool includes nine real-world scientific applications (see
The tool includes nine real-world scientific applications (see
[
benchmark table
](
https://www.spec.org/hpc2021/docs/result-fields.html#benchmarks
)
)
[
benchmark table
](
https://www.spec.org/hpc2021/docs/result-fields.html#benchmarks
)
)
with different workload sizes ranging from tiny, small, medium to large, and different parallelization models
with different workload sizes ranging from tiny, small, medium to large, and different parallelization models
including MPI only, MPI+OpenACC, MPI+OpenMP and MPI+OpenMP with target offloading. With th
e
including MPI only, MPI+OpenACC, MPI+OpenMP and MPI+OpenMP with target offloading. With th
is
benchmark you can
not only
compare the performance of different systems, it can also be used to
benchmark
suite
you can compare the performance of different
HPC
systems
. Furthermore
, it can also be used to
evaluate parallel strategies for applications on a target HPC system.
evaluate parallel strategies for applications on a target HPC system.
When you e.g. want to implement an algorithm, port an application to another platform or integrate acceleration into
When you e.g. want to implement an algorithm, port an application to another platform or integrate acceleration into
your code, you can determine from which target system and parallelization model your application
your code, you can determine from which target system and parallelization model your application
...
@@ -90,14 +90,13 @@ medium or large, test or reference).
...
@@ -90,14 +90,13 @@ medium or large, test or reference).
```
bash linenums="1"
```
bash linenums="1"
#!/bin/bash
#!/bin/bash
#SBATCH --account=<p_number_crunch>
#SBATCH --account=<p_number_crunch>
#SBATCH --
time=16:00:00
#SBATCH --
partition=haswell64
#SBATCH --exclusive
#SBATCH --exclusive
#SBATCH --nodes=1
#SBATCH --nodes=1
#SBATCH --ntasks=24
#SBATCH --ntasks=24
#SBATCH --ntasks-per-node=24
#SBATCH --cpus-per-task=1
#SBATCH --cpus-per-task=1
#SBATCH --mem-per-cpu=2541
#SBATCH --mem-per-cpu=2541
#SBATCH --
partition=haswell64
#SBATCH --
time=16:00:00
#SBATCH --constraint=DA
#SBATCH --constraint=DA
module purge
module purge
...
@@ -248,9 +247,7 @@ medium or large, test or reference).
...
@@ -248,9 +247,7 @@ medium or large, test or reference).
!!! failure "Error: system limit exceeded on number of processes that can be started"
!!! failure "Error: system limit exceeded on number of processes that can be started"
ORTE_ERROR_LOG: The system limit on number of children a process can have was reached. This can
ORTE_ERROR_LOG: The system limit on number of children a process can have was reached.
be resolved by either asking the system administrator for that node to increase the system
limit, or by rearranging your processes to place fewer of them on that node.
!!! note "Explanation"
!!! note "Explanation"
...
...
This diff is collapsed.
Click to expand it.
Preview
0%
Loading
Try again
or
attach a new file
.
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Save comment
Cancel
Please
register
or
sign in
to comment