Skip to content
4ip_extension.rst 9.01 KiB
Newer Older
Victor's avatar
Victor committed
.. _4ip-extension:

Victor's avatar
Victor committed
4IP-extension: Run on PCP machines
Victor's avatar
Victor committed
==================================

Victor's avatar
Victor committed
This document will describe the work done under the PRACE-4IP extension. This task is dedicated to provide useful information on application *performance and energy usage* on next generation systems on the path towards exacsale. It will be caried out running the accelerated UEABS on PCP systems to obtain energy metrics on *OpenPower+GPU, Xeon Phi and FPGA*.
Victor's avatar
Victor committed

Organisation is publicly available to anyone interrested in the project and rely on the following tools:
Victor's avatar
Victor committed
 * a slack channel for chat purpose: `PRACE PCP slack channel`_ , ask `Victor Cameo Ponz`_ for registration link.
Victor's avatar
Victor committed
 * a mailing list: prace-4ip-wp7.extension@fz-juelich.de, subscribe here: `mailing list registration page`_.
Victor's avatar
Victor committed
 * this documentation.
Victor's avatar
Victor committed

The following methodology will be followed to lead this task:
 #. white paper indicating what applications will be run on which prototypes (meeting the PRACE milestone **MS33 due by August 2017/M4**)
Victor's avatar
Victor committed
 #. grant access to machines (cut off September 2017 )
 #. run benchmarks (cut off October/November)
Victor's avatar
Victor committed
 #. gather results and report *Applications Performance and Energy Usage* (this adresses the PRACE deliverable **D7.7 due by December 2017/M8**)
Victor's avatar
Victor committed

Code definition
---------------

Victor's avatar
Victor committed
The table :ref:`table-code-definition` shows all codes available with UEABS (regular and accelerated). It states for each codes, tageted architechures and BCOs. Note that due to tight deadlines, efforts to port codes to new architechures will have to be minimal.
Victor's avatar
Victor committed

.. _table-code-definition:
.. table:: Code definition
   :widths: auto

   +------------------------+--------------------------------+-------------------------------+
   |                        |           Will run on          |                               |
   |   Code name            +--------------+----------+------+   4IP-extension BCO           +
   |                        | Power8 + GPU | Xeon Phi | FPGA |                               |
   +========================+==============+==========+======+===============================+
   | ALYA                   | ✓ (?)        | ✓ (?)    | ✗    | ? Ricard Borrell (BSC) ??     |
   +------------------------+--------------+----------+------+-------------------------------+
   | Code_Saturne           | ✓            | ✓        | ✗    | Charles Moulinec (STFC)       |
   +------------------------+--------------+----------+------+-------------------------------+
   | CP2K                   | ✓            | ✓        | ✗    | Arno Proeme (EPCC)            |
   +------------------------+--------------+----------+------+-------------------------------+
   | GADGET                 | ✗            | ✓        | ✗    | Volker Weinberg (LRZ)         |
   +------------------------+--------------+----------+------+-------------------------------+
   | GENE                   |              |          | ✗    |  ?                            |
   +------------------------+--------------+----------+------+-------------------------------+
   | GPAW                   |     ✗        | ✓        | ✗    | Martti Louhivuori (CINCA)     |
   +------------------------+--------------+----------+------+-------------------------------+
   | GROMACS                |  ✓           | ✓        | ✗    | Dimitris Dellis (GRNET)       |
   +------------------------+--------------+----------+------+-------------------------------+
   | NAMD                   |  ✓           | ✓        | ✗    | Dimitris Dellis (GRNET)       |
   +------------------------+--------------+----------+------+-------------------------------+
   | NEMO                   |              |          | ?    |  ?                            |
   +------------------------+--------------+----------+------+-------------------------------+
   | PFARM                  |  ✓           | ✓        | ✗    | Mariusz Uchronski (WCNS/PSNC) |
   +------------------------+--------------+----------+------+-------------------------------+
   | QCD                    |    ✓         | ✓        | ✗    | Arno Proeme (EPCC)            |
   +------------------------+--------------+----------+------+-------------------------------+
   | Quantum Espresso       |    ✓         |    ✓     | ✓    | Andrew Emerson (CINECA)       |
   +------------------------+--------------+----------+------+-------------------------------+
   | SHOC                   |    ✓         | ✗        | ✓    | Valeriu Codreanu (SurfSARA)   |
   +------------------------+--------------+----------+------+-------------------------------+
   | Specfem3D_Globe        |    ✓         | ✓        | ✓    | Victor Cameo Ponz (CINES)     |
   +------------------------+--------------+----------+------+-------------------------------+
Victor's avatar
Victor committed


PCP systems description and registration process
------------------------------------------------

Victor's avatar
Victor committed
This section describes the systems where BCOs have been granted access.
The table :ref:`table-pcp-systems` sums up the when registration process starts:
Victor's avatar
Victor committed
.. _table-pcp-systems:
.. table:: PCP Systems
Victor's avatar
Victor committed
   :widths: auto

Victor's avatar
Victor committed
   +--------------+--------------+----------------------------+---------------+-------------------------------+
   |  Technology  | Theoretical  | Manufacturer               | Host          | Availability                  |
   |              | peak perf    |                            |               |                               |
   +==============+==============+============================+===============+===============================+
   | Power8 + GPU | 877 TFlop/s  | `E4 computer engineering`_ | CINECA_ (It)  | June/July 2017                |
   |              |              |                            |               | **shifted to mid-September**  |
   +--------------+--------------+----------------------------+---------------+-------------------------------+
Victor's avatar
Victor committed
   | Xeon Phi     | 512 TFlop/s  | `Atos/Bull`_               | CINES_ (Fr)   | June 2017 (now available)     |
Victor's avatar
Victor committed
   +--------------+--------------+----------------------------+---------------+-------------------------------+
   | FPGA         | N/A          | MAXELER_                   | JSC_ (De)     | August 2017                   |
   |              |              |                            |               | **shifted to mid-September**  |
   +--------------+--------------+----------------------------+---------------+-------------------------------+
Victor's avatar
Victor committed

.. note:: The registration process included whithin is aimed at BCOs only.

Victor's avatar
Victor committed

Power8 + GPU
^^^^^^^^^^^^

Victor's avatar
Victor committed
This machine has been designed by `E4 computer engineering`_ and is hosted at CINECA_ in Bologna, Italy.

.. note:: In order to access the machine BCO should register on the `CINECA user datatabase`_ and ask `Victor Cameo Ponz`_ to be added to the 4IP-extension project.
Victor's avatar
Victor committed

Compute technology
""""""""""""""""""

Hardware features fat-nodes with the following design:
Victor's avatar
Victor committed
 * 45 nodes with x2 IBM POWER8 processors and x4 NVIDIA P100 GPU
 * intranode comunications integrated using NVLink
 * extranode comunications integrated using Infiniband ERD interconnect
Victor's avatar
Victor committed
 * CPU and GPU liquid cooling based on CoolIT_ solution

Energy sampling technology
""""""""""""""""""""""""""

Xeon Phi
^^^^^^^^

Victor's avatar
Victor committed
This machine has been designed by `Atos/Bull`_ and is hosted at CINES_ in Montpellier, France.

.. note::

    In order to access the machine BCO should fill the CINES login opening form: `odt <cines-login-form-odt_>`_ or `rtf <cines-login-form-rtf_>`_.
    Use the following information to fill project related fields:
Victor's avatar
Victor committed

     - project outside DARI
     - name of the personn in charge of the project: Victor Cameo Ponz
     - phone number: +33 (0)4 67 14 14 03
     - project code: praceknl
     - scientific machine demanded: PCP KNL cluster
Victor's avatar
Victor committed

    Then send it back to `Victor Cameo Ponz`_.
Victor's avatar
Victor committed

Compute technology
""""""""""""""""""
Hardware features the following nodes:
Victor's avatar
Victor committed
 * 168 nodes with 1x Intel Xeon Phi 7250 processor (KNL)
 * Liquid cooled nodes and PSU
 * MooseFS I/O
Victor's avatar
Victor committed

Energy sampling technology
""""""""""""""""""""""""""

Victor's avatar
Victor committed

Victor's avatar
Victor committed
FPGA
^^^^

This machine has been designed by MAXELER_ and is hosted at JSC_ in Julich, Germany.

Compute technology
""""""""""""""""""

Victor's avatar
Victor committed
This small pilot system features:
 - 4 MPC-H servers including 2x MAX5 DFE and 2x Intel Xeon processors

Victor's avatar
Victor committed
Energy sampling technology
""""""""""""""""""""""""""



Victor's avatar
Victor committed
.. _Victor Cameo Ponz: cameo+4ip-extension@cines.fr
Victor's avatar
Victor committed
.. _mailing list registration page: https://lists.fz-juelich.de/mailman/listinfo/prace-4ip-wp7.extension
Victor's avatar
Victor committed

.. _PRACE PCP slack channel: https://prace-pcp.slack.com

.. _E4 computer engineering: https://www.e4company.com
.. _Atos/Bull: https://bull.com/
.. _MAXELER: http://maxeler.com/

.. _CINECA: http://hpc.cineca.it/
.. _CINES: https://www.cines.fr/
.. _JSC: http://www.fz-juelich.de/ias/jsc/EN/Home/home_node.html

.. _CINECA user datatabase: https://userdb.hpc.cineca.it/
.. _cines-login-form-odt: https://www.cines.fr/wp-content/uploads/2014/01/opening_renewal_login_2017.odt
.. _cines-login-form-rtf: https://www.cines.fr/wp-content/uploads/2014/01/opening_renewal_login_2017.rtf
Victor's avatar
Victor committed

.. _CoolIT: https://www.coolitsystems.com/
.. _Slurm: https://slurm.schedmd.com/