Rennes:Hardware: Difference between revisions
Jump to navigation
Jump to search
Jlelaurain (talk | contribs) No edit summary |
No edit summary |
||
Line 6: | Line 6: | ||
= Summary = | = Summary = | ||
* 5 clusters | * 5 clusters | ||
* | * 111 nodes | ||
* | * 2044 CPU cores | ||
* | * 11 GPUs | ||
* | * 15.75 TiB RAM | ||
* | * 39 SSDs and 316 HDDs on nodes (total: 207.84 TB) | ||
* | * 76.8 TFLOPS (excluding GPUs) | ||
= Clusters = | = Clusters = | ||
Line 32: | Line 32: | ||
|- | |- | ||
|[[#abacus25|abacus25]]||<b>[[Grid5000:UsagePolicy#Rules_for_the_production_queue|production]]</b> queue||2023-06-05|| | |[[#abacus25|abacus25]]||<b>[[Grid5000:UsagePolicy#Rules_for_the_production_queue|production]]</b> queue||2023-06-05||4||2||AMD EPYC 7413||24 cores/CPU||x86_64||512 GiB||data-sort-value="2682"|<b>960 GB SSD</b> + 2 x 960 GB SSD||data-sort-value="10000"|10 Gbps (SR‑IOV) ||[1-3]: 3 x Nvidia A40 (45 GiB)<br />4: 2 x Nvidia A40 (45 GiB) | ||
|- | |- | ||
|[[#paranoia|paranoia]]||<b>[[Grid5000:UsagePolicy#Rules_for_the_production_queue|production]]</b> queue||2014-02-21||7||2||Intel Xeon E5-2660 v2||10 cores/CPU||x86_64||128 GiB||data-sort-value="2790"|<b>600 GB HDD</b> + 4 x 600 GB HDD||data-sort-value="21000"|1 Gbps (SR‑IOV) + 2 x 10 Gbps (SR‑IOV) || | |[[#paranoia|paranoia]]||<b>[[Grid5000:UsagePolicy#Rules_for_the_production_queue|production]]</b> queue||2014-02-21||7||2||Intel Xeon E5-2660 v2||10 cores/CPU||x86_64||128 GiB||data-sort-value="2790"|<b>600 GB HDD</b> + 4 x 600 GB HDD||data-sort-value="21000"|1 Gbps (SR‑IOV) + 2 x 10 Gbps (SR‑IOV) || | ||
Line 231: | Line 231: | ||
== [https://intranet.grid5000.fr/oar/Rennes/drawgantt-svg-prod/?filter=abacus25%20only abacus25] == | == [https://intranet.grid5000.fr/oar/Rennes/drawgantt-svg-prod/?filter=abacus25%20only abacus25] == | ||
''' | '''4 nodes, 8 cpus, 192 cores,''' split as follows due to differences between nodes ([https://public-api.grid5000.fr/stable/sites/rennes/clusters/abacus25/nodes.json?pretty=1 json]) | ||
'''Reservation example:''' | '''Reservation example:''' | ||
Line 239: | Line 239: | ||
'''Max walltime per nodes:''' | '''Max walltime per nodes:''' | ||
* abacus25-1: 96h | * abacus25-1: 96h | ||
* abacus25-2: 168h | * abacus25-[2-4]: 168h | ||
; abacus25-[1-2] (2 nodes, 4 cpus, 96 cores) | |||
{| | {| | ||
|- | |- | ||
Line 271: | Line 272: | ||
| valign="top" style="background-color: #f9f9f9; padding: 0px 10px 0px 3px;" |'''GPU:''' | | valign="top" style="background-color: #f9f9f9; padding: 0px 10px 0px 3px;" |'''GPU:''' | ||
| 3 x Nvidia A40 (45 GiB)<br>Compute capability: 8.6<br/> | | 3 x Nvidia A40 (45 GiB)<br>Compute capability: 8.6<br/> | ||
|- | |||
|} | |||
<hr style="height:10pt; visibility:hidden;" /> | |||
; abacus25-3 (1 node, 2 cpus, 48 cores) | |||
{| | |||
|- | |||
| valign="top" style="background-color: #f9f9f9; padding: 0px 10px 0px 3px;" |'''Access condition:''' | |||
| production queue<br/> | |||
|- | |||
| valign="top" style="background-color: #f9f9f9; padding: 0px 10px 0px 3px;" |'''Model:''' | |||
| HPE ProLiant DL385 Gen10+ v2<br/> | |||
|- | |||
| valign="top" style="background-color: #f9f9f9; padding: 0px 10px 0px 3px;" |'''Date of arrival:''' | |||
| 2023-06-05<br/> | |||
|- | |||
| valign="top" style="background-color: #f9f9f9; padding: 0px 10px 0px 3px;" |'''CPU:''' | |||
| AMD EPYC 7413 (Zen 3), x86_64, 2 CPUs/node, 24 cores/CPU<br/> | |||
|- | |||
| valign="top" style="background-color: #f9f9f9; padding: 0px 10px 0px 3px;" |'''Memory:''' | |||
| 512 GiB<br/> | |||
|- | |||
| valign="top" style="background-color: #f9f9f9; padding: 0px 10px 0px 3px;" |'''Storage:''' | |||
| | |||
* disk0, 960 GB SSD SAS HPE MO000960RXRQK (dev: <code class="file">/dev/disk0</code>, by-path: <code class="file">/dev/disk/by-path/pci-0000:64:00.0-scsi-0:2:1:0</code>) (primary disk)<br /> | |||
* disk1, 960 GB SSD SATA HP VK000960GZCNF (dev: <code class="file">/dev/disk1</code>, by-path: <code class="file">/dev/disk/by-path/pci-0000:64:00.0-scsi-0:2:2:0</code>) <br /> | |||
* disk2, 960 GB SSD SATA HP VK000960GZCNF (dev: <code class="file">/dev/disk2</code>, by-path: <code class="file">/dev/disk/by-path/pci-0000:64:00.0-scsi-0:2:3:0</code>) <br/> | |||
|- | |||
| valign="top" style="background-color: #f9f9f9; padding: 0px 10px 0px 3px;" |'''Network:''' | |||
| | |||
* eth0/ens10f0, Ethernet, configured rate: 10 Gbps, model: Intel Ethernet Controller X710 for 10GbE SFP+, driver: i40e, SR-IOV enabled - no KaVLAN<br /> | |||
* <span style="color:grey">eth1/ens10f1, Ethernet, model: Intel Ethernet Controller X710 for 10GbE SFP+, driver: i40e - unavailable for experiment</span><br/> | |||
|- | |||
| valign="top" style="background-color: #f9f9f9; padding: 0px 10px 0px 3px;" |'''GPU:''' | |||
| 3 x Nvidia A40 (45 GiB)<br>Compute capability: 8.6<br/> | |||
|- | |||
|} | |||
<hr style="height:10pt; visibility:hidden;" /> | |||
; abacus25-4 (1 node, 2 cpus, 48 cores) | |||
{| | |||
|- | |||
| valign="top" style="background-color: #f9f9f9; padding: 0px 10px 0px 3px;" |'''Access condition:''' | |||
| production queue<br/> | |||
|- | |||
| valign="top" style="background-color: #f9f9f9; padding: 0px 10px 0px 3px;" |'''Model:''' | |||
| HPE ProLiant DL385 Gen10+ v2<br/> | |||
|- | |||
| valign="top" style="background-color: #f9f9f9; padding: 0px 10px 0px 3px;" |'''Date of arrival:''' | |||
| 2023-06-05<br/> | |||
|- | |||
| valign="top" style="background-color: #f9f9f9; padding: 0px 10px 0px 3px;" |'''CPU:''' | |||
| AMD EPYC 7413 (Zen 3), x86_64, 2 CPUs/node, 24 cores/CPU<br/> | |||
|- | |||
| valign="top" style="background-color: #f9f9f9; padding: 0px 10px 0px 3px;" |'''Memory:''' | |||
| 512 GiB<br/> | |||
|- | |||
| valign="top" style="background-color: #f9f9f9; padding: 0px 10px 0px 3px;" |'''Storage:''' | |||
| | |||
* disk0, 960 GB SSD SAS HPE MO000960RXRQK (dev: <code class="file">/dev/disk0</code>, by-path: <code class="file">/dev/disk/by-path/pci-0000:64:00.0-scsi-0:2:1:0</code>) (primary disk)<br /> | |||
* disk1, 960 GB SSD SATA HP VK000960GXNZB (dev: <code class="file">/dev/disk1</code>, by-path: <code class="file">/dev/disk/by-path/pci-0000:64:00.0-scsi-0:2:2:0</code>) <br /> | |||
* disk2, 960 GB SSD SATA HP VK000960GXNZB (dev: <code class="file">/dev/disk2</code>, by-path: <code class="file">/dev/disk/by-path/pci-0000:64:00.0-scsi-0:2:3:0</code>) <br/> | |||
|- | |||
| valign="top" style="background-color: #f9f9f9; padding: 0px 10px 0px 3px;" |'''Network:''' | |||
| | |||
* eth0/ens10f0, Ethernet, configured rate: 10 Gbps, model: Intel Ethernet Controller X710 for 10GbE SFP+, driver: i40e, SR-IOV enabled - no KaVLAN<br /> | |||
* <span style="color:grey">eth1/ens10f1, Ethernet, model: Intel Ethernet Controller X710 for 10GbE SFP+, driver: i40e - unavailable for experiment</span><br/> | |||
|- | |||
| valign="top" style="background-color: #f9f9f9; padding: 0px 10px 0px 3px;" |'''GPU:''' | |||
| 2 x Nvidia A40 (45 GiB)<br>Compute capability: 8.6<br/> | |||
|- | |- | ||
|} | |} | ||
Line 469: | Line 540: | ||
* <span style="color:grey">eth3/eno34, Ethernet, model: Intel Ethernet Controller X710 for 10GbE SFP+, driver: i40e - unavailable for experiment</span><br/> | * <span style="color:grey">eth3/eno34, Ethernet, model: Intel Ethernet Controller X710 for 10GbE SFP+, driver: i40e - unavailable for experiment</span><br/> | ||
|- | |- | ||
|}''<small>Last generated from the Grid'5000 Reference API on 2023-09- | |}''<small>Last generated from the Grid'5000 Reference API on 2023-09-26 ([https://gitlab.inria.fr/grid5000/reference-repository/commit/000d109311 commit 000d109311])</small>'' |
Revision as of 09:09, 26 September 2023
Hardware: Global | Grenoble | Lille | Luxembourg | Lyon | Nancy | Nantes | Rennes | Sophia | Toulouse
See also: Network topology for Rennes
Summary
- 5 clusters
- 111 nodes
- 2044 CPU cores
- 11 GPUs
- 15.75 TiB RAM
- 39 SSDs and 316 HDDs on nodes (total: 207.84 TB)
- 76.8 TFLOPS (excluding GPUs)
Clusters
Cluster | Access Condition | Date of arrival | Nodes | CPU | Memory | Storage | Network | Accelerators | |||
---|---|---|---|---|---|---|---|---|---|---|---|
# | Name | Cores | Architecture | ||||||||
abacus25 | production queue | 2023-06-05 | 4 | 2 | AMD EPYC 7413 | 24 cores/CPU | x86_64 | 512 GiB | 960 GB SSD + 2 x 960 GB SSD | 10 Gbps (SR‑IOV) | [1-3]: 3 x Nvidia A40 (45 GiB) 4: 2 x Nvidia A40 (45 GiB) |
paranoia | production queue | 2014-02-21 | 7 | 2 | Intel Xeon E5-2660 v2 | 10 cores/CPU | x86_64 | 128 GiB | 600 GB HDD + 4 x 600 GB HDD | 1 Gbps (SR‑IOV) + 2 x 10 Gbps (SR‑IOV) | |
parasilo | 2015-01-13 | 27 | 2 | Intel Xeon E5-2630 v3 | 8 cores/CPU | x86_64 | 128 GiB | 600 GB HDD + 4 x 600 GB HDD* + 200 GB SSD* | 2 x 10 Gbps (SR‑IOV) | ||
paravance | 2015-01-13 | 72 | 2 | Intel Xeon E5-2630 v3 | 8 cores/CPU | x86_64 | 128 GiB | 600 GB HDD + 600 GB HDD | 2 x 10 Gbps (SR‑IOV) | ||
roazhon4 | production queue | 2023-09-06 | 1 | 2 | AMD EPYC 7H12 | 64 cores/CPU | x86_64 | 512 GiB | 480 GB HDD + 2.0 TB HDD | 1 Gbps |
*: disk is reservable
**: crossed GPUs are not supported by Grid'5000 default environments
Clusters in the default queue
parasilo
27 nodes, 54 cpus, 432 cores, split as follows due to differences between nodes (json)
Reservation example:
- parasilo-[1-5,
7-12, 14-15, 17-24, 26-28] (24 nodes, 48 cpus, 384 cores)
Model: | Dell PowerEdge R630 |
Date of arrival: | 2015-01-13 |
CPU: | Intel Xeon E5-2630 v3 (Haswell), x86_64, 2.40GHz, 2 CPUs/node, 8 cores/CPU |
Memory: | 128 GiB |
Storage: |
|
Network: |
|
- parasilo-6 (1 node, 2 cpus, 16 cores)
Model: | Dell PowerEdge R630 |
Date of arrival: | 2015-01-13 |
CPU: | Intel Xeon E5-2630 v3 (Haswell), x86_64, 2.40GHz, 2 CPUs/node, 8 cores/CPU |
Memory: | 128 GiB |
Storage: |
|
Network: |
|
- parasilo-13 (1 node, 2 cpus, 16 cores)
Model: | Dell PowerEdge R630 |
Date of arrival: | 2015-01-13 |
CPU: | Intel Xeon E5-2630 v3 (Haswell), x86_64, 2.40GHz, 2 CPUs/node, 8 cores/CPU |
Memory: | 128 GiB |
Storage: |
|
Network: |
|
- parasilo-25 (1 node, 2 cpus, 16 cores)
Model: | Dell PowerEdge R630 |
Date of arrival: | 2015-01-13 |
CPU: | Intel Xeon E5-2630 v3 (Haswell), x86_64, 2.40GHz, 2 CPUs/node, 8 cores/CPU |
Memory: | 128 GiB |
Storage: |
|
Network: |
|
paravance
72 nodes, 144 cpus, 1152 cores (json)
Reservation example:
Model: | Dell PowerEdge R630 |
Date of arrival: | 2015-01-13 |
CPU: | Intel Xeon E5-2630 v3 (Haswell), x86_64, 2.40GHz, 2 CPUs/node, 8 cores/CPU |
Memory: | 128 GiB |
Storage: |
|
Network: |
|
Clusters in the production queue
abacus25
4 nodes, 8 cpus, 192 cores, split as follows due to differences between nodes (json)
Reservation example:
Max walltime per nodes:
- abacus25-1: 96h
- abacus25-[2-4]: 168h
- abacus25-[1-2] (2 nodes, 4 cpus, 96 cores)
Access condition: | production queue |
Model: | HPE ProLiant DL385 Gen10+ v2 |
Date of arrival: | 2023-06-05 |
CPU: | AMD EPYC 7413 (Zen 3), x86_64, 2 CPUs/node, 24 cores/CPU |
Memory: | 512 GiB |
Storage: |
|
Network: |
|
GPU: | 3 x Nvidia A40 (45 GiB) Compute capability: 8.6 |
- abacus25-3 (1 node, 2 cpus, 48 cores)
Access condition: | production queue |
Model: | HPE ProLiant DL385 Gen10+ v2 |
Date of arrival: | 2023-06-05 |
CPU: | AMD EPYC 7413 (Zen 3), x86_64, 2 CPUs/node, 24 cores/CPU |
Memory: | 512 GiB |
Storage: |
|
Network: |
|
GPU: | 3 x Nvidia A40 (45 GiB) Compute capability: 8.6 |
- abacus25-4 (1 node, 2 cpus, 48 cores)
Access condition: | production queue |
Model: | HPE ProLiant DL385 Gen10+ v2 |
Date of arrival: | 2023-06-05 |
CPU: | AMD EPYC 7413 (Zen 3), x86_64, 2 CPUs/node, 24 cores/CPU |
Memory: | 512 GiB |
Storage: |
|
Network: |
|
GPU: | 2 x Nvidia A40 (45 GiB) Compute capability: 8.6 |
paranoia
7 nodes, 14 cpus, 140 cores, split as follows due to differences between nodes (json)
Reservation example:
Max walltime per nodes:
- paranoia-[1-4]: 12h
- paranoia-[5-7]: 168h
- paranoia-[1-2,
5, 7] (4 nodes, 8 cpus, 80 cores)
Access condition: | production queue |
Model: | Dell PowerEdge C6220 II |
Date of arrival: | 2014-02-21 |
CPU: | Intel Xeon E5-2660 v2 (Sandy Bridge), x86_64, 2.20GHz, 2 CPUs/node, 10 cores/CPU |
Memory: | 128 GiB |
Storage: |
|
Network: |
|
- paranoia-3 (1 node, 2 cpus, 20 cores)
Access condition: | production queue |
Model: | Dell PowerEdge C6220 II |
Date of arrival: | 2014-02-21 |
CPU: | Intel Xeon E5-2660 v2 (Sandy Bridge), x86_64, 2.20GHz, 2 CPUs/node, 10 cores/CPU |
Memory: | 128 GiB |
Storage: |
|
Network: |
|
- paranoia-4 (1 node, 2 cpus, 20 cores)
Access condition: | production queue |
Model: | Dell PowerEdge C6220 II |
Date of arrival: | 2014-02-21 |
CPU: | Intel Xeon E5-2660 v2 (Sandy Bridge), x86_64, 2.20GHz, 2 CPUs/node, 10 cores/CPU |
Memory: | 128 GiB |
Storage: |
|
Network: |
|
- paranoia-8 (1 node, 2 cpus, 20 cores)
Access condition: | production queue |
Model: | Dell PowerEdge C6220 II |
Date of arrival: | 2014-02-21 |
CPU: | Intel Xeon E5-2660 v2 (Sandy Bridge), x86_64, 2.20GHz, 2 CPUs/node, 10 cores/CPU |
Memory: | 128 GiB |
Storage: |
|
Network: |
|
roazhon4
1 node, 2 cpus, 128 cores (json)
Reservation example:
Max walltime per nodes:
- roazhon4-1: 168h
Access condition: | production queue |
Model: | Dell PowerEdge R7525 |
Date of arrival: | 2023-09-06 |
CPU: | AMD EPYC 7H12 (Zen 2), x86_64, 2 CPUs/node, 64 cores/CPU |
Memory: | 512 GiB |
Storage: |
|
Network: |
|
Last generated from the Grid'5000 Reference API on 2023-09-26 (commit 000d109311)