Mp2/en: Difference between revisions

From Alliance Doc
Jump to navigation Jump to search
No edit summary
No edit summary
 
(8 intermediate revisions by 3 users not shown)
Line 2: Line 2:
{| class="wikitable"
{| class="wikitable"
|-
|-
| Availability: February 2012 (in operation until April 1, 2020)
| Availability: February 2012 - April 1, 2020
|-
|-
| Login Node: '''mp2.calculcanada.ca'''
| Login Node: '''mp2.calculcanada.ca'''
Line 11: Line 11:
|}
|}


Mammouth-Mp2 is a heterogeneous and polyvalent cluster designed for ordinary computations; it is located at the [http://www.usherbrooke.ca/ Université de Sherbrooke]. It is a legacy resource which is administered by the regional consortium Calcul Québec. It is not part of the new national clusters of Compute Canada. 
'''Mp2 is now exclusive to researchers from the Université de Sherbrooke.'''
 
Mammouth-Mp2 is a heterogeneous and polyvalent cluster designed for ordinary computations; it is located at the [http://www.usherbrooke.ca/ Université de Sherbrooke].


=Site-specific policies=
=Site-specific policies=
Line 32: Line 34:
| SCRATCH <br> Lustre filesystem <br> 358.3 TB  of space in total ||
| SCRATCH <br> Lustre filesystem <br> 358.3 TB  of space in total ||


*Grand espace pour stocker les fichiers temporaires pendant les calculs.
*Large space for storing temporary files during computations.


*20 TB of space and 1M files per user.  
*20 TB of space and 1M files per user.  
Line 41: Line 43:
| PROJECT <br> Lustre filesystem <br> 716.6 TB of space in total ||
| PROJECT <br> Lustre filesystem <br> 716.6 TB of space in total ||


*Cet espace est conçu pour le partage de données entre membres d'un groupe et pour le stockage de beaucoup de données.  
*This space is designed for sharing data among the members of a research group and for storing large amounts of data.  


*1 TB  of space and 500K files per group.  
*1 TB  of space and 500K files per group.  
Line 48: Line 50:
|}
|}


Pour les transferts de données par Globus, on devrait utiliser le point de chute <code>computecanada#mammouth</code>, alors que pour les outils comme rsync et scp, on peut utiliser un nœud
For transferring data by Globus, you should use the endpoint <code>computecanada#mammouth</code>, whereas tools like rsync and scp can simply use an ordinary login node.
de connexion.


=High-performance interconnect=
=High-performance interconnect=


Le réseau Infiniband QDR (40 Gb/s) de Mellanox relie tous les nœuds de la grappe, non-bloquant sur 216 noeds, 5:1 pour le reste.
The Mellanox QDR (40 Gb/s) Infiniband network links together all of the cluster's nodes and is non-blocking for groups of 216 nodes and 5:1 for the rest of the cluster.


=Node characteristics=
=Node characteristics=

Latest revision as of 21:58, 22 April 2021

Other languages:
Availability: February 2012 - April 1, 2020
Login Node: mp2.calculcanada.ca
Globus Endpoint: computecanada#mammouth
Data Transfer Node (rsync, scp, sftp,...) : mp2.calculcanada.ca

Mp2 is now exclusive to researchers from the Université de Sherbrooke.

Mammouth-Mp2 is a heterogeneous and polyvalent cluster designed for ordinary computations; it is located at the Université de Sherbrooke.

Site-specific policies

  • Each job must have a duration of at least one hour (at least five minutes for test jobs) and a user cannot have more than 1000 jobs (running and queued) at any given time. The maximum duration of a job is 168 hours (seven days).
  • No GPUs.

Storage

HOME
Lustre filesystem
79.6 TB of space in total
  • This space is small and cannot be expanded; you should use your project space for substantial storage needs.
  • 50 GB of space and 500K files per user.
  • There is a daily backup.
SCRATCH
Lustre filesystem
358.3 TB of space in total
  • Large space for storing temporary files during computations.
  • 20 TB of space and 1M files per user.
  • No backup system in place.
PROJECT
Lustre filesystem
716.6 TB of space in total
  • This space is designed for sharing data among the members of a research group and for storing large amounts of data.
  • 1 TB of space and 500K files per group.
  • No backup system in place.

For transferring data by Globus, you should use the endpoint computecanada#mammouth, whereas tools like rsync and scp can simply use an ordinary login node.

High-performance interconnect

The Mellanox QDR (40 Gb/s) Infiniband network links together all of the cluster's nodes and is non-blocking for groups of 216 nodes and 5:1 for the rest of the cluster.

Node characteristics

Quantity Cores Available Memory CPU Type Storage GPU Type
1588 24 31 GB or 31744 MB 12 cores/socket, 2 sockets/node. AMD Opteron Processor 6172 @ 2.1 GHz 1TB SATA disk. -
20 48 251 GB or 257024 MB 12 cores/socket, 4 sockets/node. AMD Opteron Processor 6174 @ 2.2 GHz 1TB SATA disk. -
2 48 503 GB or 515072 MB 12 cores/socket, 4 sockets/node. AMD Opteron Processor 6174 @ 2.2 GHz 1TB SATA disk. -