Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision		 Previous revision
habrok:introduction:cluster_description [2024/11/22 08:16] – fokkehabrok:introduction:cluster_description [2026/02/27 10:14] (current) – fokke
Line 8: Line 8:
 {{ :habrok:introduction:habrok_standard_compute_node.jpg?direct&150|Standard Hábrók compute node}} {{ :habrok:introduction:habrok_standard_compute_node.jpg?direct&150|Standard Hábrók compute node}}
  
-  * 119 standard nodes with the following components:+  * 84 standard nodes with the following components:
     * 128 cores @ 2.45 GHz (two AMD 7763 CPUs)     * 128 cores @ 2.45 GHz (two AMD 7763 CPUs)
     * 512 GB memory     * 512 GB memory
     * 3.5 TB internal SSD disk space     * 3.5 TB internal SSD disk space
  
-  * 24 nodes for multi-node jobs with the following components:+  * 48 nodes for multi-node jobs with the following components:
     * 128 cores @ 2.45 GHz (two AMD 7763 CPUs)     * 128 cores @ 2.45 GHz (two AMD 7763 CPUs)
     * 512 GB memory     * 512 GB memory
Line 25: Line 25:
     * 4096 GB memory     * 4096 GB memory
     * 14 TB internal SSD disk space     * 14 TB internal SSD disk space
 +
 +  * 2 Interactive GPU nodes (Delivered by Fujitsu in an earlier purchase) with the following components:
 +    * 24 cores @ 2.4 GHz (two Intel Xeon Gold 6240R CPUs)
 +    * 768 GB memory
 +    * 1 Nvidia L40s GPU accelerator card with 48GB RAM
  
   * 6 GPU nodes with the following components:   * 6 GPU nodes with the following components:
Line 33: Line 38:
     * 100 Gbps Omni-Path link     * 100 Gbps Omni-Path link
 {{ :habrok:introduction:habrok_power_nodes.jpg?direct&150|Hábrók nodes with cables for power and network}} {{ :habrok:introduction:habrok_power_nodes.jpg?direct&150|Hábrók nodes with cables for power and network}}
 +
 +  * 3 GPU nodes with the following components:
 +    * 128 cores @ 3.3 GHz (two AMD EPYC 9575F CPUs)
 +    * 1536 GB memory
 +    * 8 Nvidia RTX Pro 6000 GPUs with 96 GB RAM
 +    * 15 TB internal SSD NVMe disk space
 +    * 100 Gbps Ethernet
  
   * 18 GPU nodes (Delivered by Fujitsu in an earlier purchase) with the following components:   * 18 GPU nodes (Delivered by Fujitsu in an earlier purchase) with the following components:
-    * 18 cores @ 2.7 GHz (two Intel Xeon Gold 6150 CPUs)+    * 36 cores @ 2.7 GHz (two Intel Xeon Gold 6150 CPUs)
     * 768 GB memory (621 GB used for temporary disk space)     * 768 GB memory (621 GB used for temporary disk space)
-    * 1 Nvidia V100 GPU accelerator card with 32 GB RAM+    * 2 Nvidia V100 GPU accelerator cards each with 32 GB RAM
     * 621 GB RAM disk     * 621 GB RAM disk
  
-  * 15 nodes with the following components: +  * 1 node with the following components: 
-    * 64 cores @ 2.2 GHz (two AMD EPYC 7601 CPUs) +    * 64 cores @ 2.1 GHz (two Intel Xeon Gold 6448Y CPUs) 
-    * 512 GB memory +    * 1 TB memory 
-    * 16 TB internal disk space +    * 440 GB internal disk space 
-    * Only accessible by GELIFES users, see [[habrok:job_management:partition_details:gelifes|GELIFES Partition]] +    * 4 Nvidia H100 GPU accelerator cards with 80 GB RAM 
- +    * Only accessible for education purposes in the scope of the [[https://myuniversity.rug.nl/infonet/medewerkers/fse/education/who-is-who/sse-teachers-and-x-lab-teams/digital-lab|Digital Lab project (employee login required)]] \\ see also [[habrok:advanced_job_management:special_partitions|Group specific partition]]
-  * 2 Interactive GPU nodes (Delivered by Fujitsu in an earlier purchase) with the following components: +
-    * 24 cores @ 2.4 GHz (two Intel Xeon Gold 6240R CPUs) +
-    * 768 GB memory +
-    * 1 Nvidia L40s GPU accelerator card with 48GB RAM+
  
 +  * 1 node with the following components:
 +    * 56 cores @ 2.0 GHz (two Intel Xeon Gold 5420+ CPUs)
 +    * 512 GB memory
 +    * 3 TB internal disk space
 +    * 2 Nvidia L40S GPU accelerator cards with 48 GB RAM
 +    * Only accessible for longer jobs, and with higher priority, by users from Computer-Aided Organic Synthesis, see [[habrok:advanced_job_management:special_partitions|Group specific partition]]
  
 ===== Network ===== ===== Network =====
Line 56: Line 70:
 {{ :habrok:introduction:habrok_network.jpg?direct&150|Hábrók network switches}} {{ :habrok:introduction:habrok_network.jpg?direct&150|Hábrók network switches}}
  
-  * A 100 Gbps low-latency non-blocking Omni-Path network for 24 compute and 6 GPU nodes+  * A 100 Gbps low-latency non-blocking Omni-Path network for 48 compute and 6 GPU nodes
     * High bandwidth (100 Gigabit per second)     * High bandwidth (100 Gigabit per second)
     * Low latency (few microseconds delay before a client starts receiving the message)     * Low latency (few microseconds delay before a client starts receiving the message)