Commit 765f1925 authored by Victor's avatar Victor
Browse files

IMPROVE machines sections

parent b9bc7bb7
......@@ -16,9 +16,9 @@ Clusters specifications and access
Access to machines
^^^^^^^^^^^^^^^^^^
.. include:: /e4_gpu.rst
.. include:: /pcp_systems/e4_gpu.rst
.. include:: /atos_knl.rst
.. include:: /pcp_systems/atos_knl.rst
Performances and energy metrics of EUABS on PCP systems
......
......@@ -2,11 +2,11 @@ PCP systems
***********
.. _e4_gpu:
.. include:: /e4_gpu.rst
.. include:: /pcp_systems/e4_gpu.rst
.. _atos_knl:
.. include:: /atos_knl.rst
.. include:: /pcp_systems/atos_knl.rst
.. _maxeler_fpga:
.. include:: /maxeler_fpga.rst
.. include:: /pcp_systems/maxeler_fpga.rst
......@@ -20,6 +20,7 @@ Compute technology
""""""""""""""""""
Hardware features the following nodes:
* 168 nodes with
* 1x Intel Xeon Phi 7250 processor (KNL), 68 cores cadenced to 1.4 GHz with SMT 4.
* 96GB memory, 16GBx6 DDR4 DIMMs
* intranode comunications integrated using InfiniBand EDR
......@@ -36,6 +37,10 @@ Power measurements at node level occurs at the sampling rate of 1 kHz at convert
`Atos/Bull`_ allow energy access through two frameworks, namely HDEEM VIZualization (HDEEVIZ) and Bull Energy Optimizer (BEO).
.. note::
Specific setup documentations and instructions is available on the machine: :code:`ls /opt/software/frioul/documentation/`.
HDEEVIZ
-------
......@@ -59,6 +64,18 @@ Here's an example of usage in a submission script:
hdeeviz mpirun -n 89 $PWD/bin/xspecfem3D
Access to generated data will be made through the Grafana web interface:
.. image:: /pcp_systems/graphana.png
BEO
---
BEO is an admin oriented tools that allow to get energy metrics at switch and node level. At user level the main interesting feature is the :code:`get_job_energy <job_id<optionnal: .jobstep>>`. It produces the following output:
.. literalinclude:: /pcp_systems/output_beo_report_energy
:emphasize-lines: 1
.. _GENCI login opening form: https://www-dcc.extra.cea.fr/CCFR/
.. _cines-login-form-odt: https://www.cines.fr/wp-content/uploads/2014/01/opening_renewal_login_2017.odt
......
......@@ -10,6 +10,7 @@ Compute technology
Hardware features fat-nodes with the following design:
* 45 nodes with
* x2 IBM POWER8+ processors, ie 8x2 cores with Simultaneous Multi-Threading (SMT) 8
* x4 NVIDIA P100 GPU with 16GB High Bandwidth Memory 2 (HBM2)
* intranode comunications integrated using NVLink
......@@ -27,7 +28,7 @@ Information is collected from processors, memory, GPUs and fans exploiting Anali
The technology has been developed in collaboration with the University of Bologna which developed the :code:`get_job_energy <job_id>` program. Usage is straight forward and has the following verbose output:
.. literalinclude:: /output_get_job_energy
.. literalinclude:: /pcp_systems/output_get_job_energy
:emphasize-lines: 1
......
$ beo report energy slurm8170
| job | state | nodes.energy(slurm) | nodes.energy | switches.energy | disk_arrays.energy | job.energy | job.cost |
=============================================================================================================================
| slurm8170 | COMPLETED | | 618.4 kJ | 56.3 kJ | 0.0 J | 674.7 kJ | 0.0219 € |
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment