- Infos im HLRS Wiki sind nicht rechtsverbindlich und ohne Gewähr -
- Information contained in the HLRS Wiki is not legally binding and HLRS is not responsible for any damages that might result from its use -

NEC Cluster Hardware and Architecture (vulcan): Difference between revisions

From HLRS Platforms
Jump to navigationJump to search
No edit summary
No edit summary
Line 1: Line 1:
__TOC__
__TOC__
=== Hardware ===
=== Hardware ===


* ''' Pre- & Postprocessing node''' (''smp'' node) [Entgeltordnung: 'smp Knoten']
* ''' Pre- & Postprocessing node''' (''smp'' node) [Entgeltordnung: 'smp Knoten']
Line 20: Line 19:
*** 384GB memory
*** 384GB memory
*** 1.8TB NVMe mounted at /localscratch
*** 1.8TB NVMe mounted at /localscratch
* '''CascadeLake 36 cores compute nodes''' (''clx-ai'') for artificial intelligence and big data applications [Entgeltordnung: 'CS Storm']
** 4 nodes (''clx768gb36c-ai'')
*** 2x Intel [https://ark.intel.com/content/www/us/en/ark/products/192443/intel-xeon-gold-6240-processor-24-75m-cache-2-60-ghz.html Xeon Gold 6240], 36 cores total @ 2.60GHz
*** 768GB memory
*** 8x Nvidia Tesla V100 SXM2 32GB
*** 7.3TB NVMe mounted at /localscratch
*** 220GB SSD mounted at /tmp


* '''Haswell 24 Cores compute nodes''' (''hsw'') (only available on vulcan2)
* '''Haswell 24 Cores compute nodes''' (''hsw'') (only available on vulcan2)
Line 34: Line 24:
** 192 nodes (''hsw128gb24c'') [Entgeltordnung: 'Haswell 128 GB']  
** 192 nodes (''hsw128gb24c'') [Entgeltordnung: 'Haswell 128 GB']  
*** 128GB memory
*** 128GB memory


* '''Skylake 40 Cores compute nodes''' (''skl'') [Entgeltordnung: 'SKL-192GB']
* '''Skylake 40 Cores compute nodes''' (''skl'') [Entgeltordnung: 'SKL-192GB']
Line 40: Line 29:
*** 2x Intel [https://www.intel.com/content/www/us/en/products/processors/xeon/scalable/gold-processors/gold-6138.html Xeon Gold 6138], 40 cores total | 2.00GHz
*** 2x Intel [https://www.intel.com/content/www/us/en/products/processors/xeon/scalable/gold-processors/gold-6138.html Xeon Gold 6138], 40 cores total | 2.00GHz
*** 192GB memory
*** 192GB memory


* '''AMD 64 Cores compute nodes''' (''genoa'')
* '''AMD 64 Cores compute nodes''' (''genoa'')
Line 46: Line 34:
*** 2x AMD Epyc 9334 Genoa, 2.7GHz base, 64 cores total
*** 2x AMD Epyc 9334 Genoa, 2.7GHz base, 64 cores total
*** 768GB DDR5 memory
*** 768GB DDR5 memory


* '''AMD 64 Cores SMP nodes''' ('''genoa-smp''')
* '''AMD 64 Cores SMP nodes''' ('''genoa-smp''')
Line 52: Line 39:
*** 2x AMD Epyc 9334 Genoa, 2.7GHz base, 64 cores total
*** 2x AMD Epyc 9334 Genoa, 2.7GHz base, 64 cores total
*** 3072GB DDR5 memory
*** 3072GB DDR5 memory


*'''Visualisation node with GPUs'''
*'''Visualisation node with GPUs'''
Line 70: Line 55:
*** 3.7TB SSD mounted at /localscratch
*** 3.7TB SSD mounted at /localscratch
*** 400GB SSD mounted at /tmp
*** 400GB SSD mounted at /tmp


* '''AMD GPU nodes''' (''genoa-a30'')
* '''AMD GPU nodes''' (''genoa-a30'')
Line 78: Line 62:
*** 3.8TB NVMe mounted at /localscratch
*** 3.8TB NVMe mounted at /localscratch
*** Nvidia A30 with 24GB HBM2e
*** Nvidia A30 with 24GB HBM2e


* '''Interconnect''': [http://de.wikipedia.org/wiki/Infiniband InfiniBand]
* '''Interconnect''': [http://de.wikipedia.org/wiki/Infiniband InfiniBand]
** Various generations of Infiniband switches with QDR, FDR, EDR and HDR speed
** HDR switches (200Gbit/s); 3 spine switches; each edge switch is connected to each spine switch with 4 cables
** HDR100 connection to servers (100Gbit/s)
** skl192gb40c nodes are connected to EDR switches (100Gbit/s)


=== Architecture ===
=== Architecture ===
Line 88: Line 73:


'''Compute node types installed:'''  
'''Compute node types installed:'''  
* Haswell, Skylake, CascadeLake
* Intel Xeon Broadwell, Skylake, CascadeLake
* different Memory nodes (128GB, 256GB, 384GB)
* AMD Epyc Rome, Genoa
* Pre-Postprocessing node with very large memory (1.5TB)
* different Memory sizes (256GB, 384GB, 512GB, 768GB)
* Visualisation/GPU nodes with AMD Radeon Pro WX8200, Nvidia Quadro RTX4000 or Nvidia Tesla P100
* Pre-Postprocessing node with very large memory (1.5TB, 3TB)
* KI nodes with Nvidia Tesla V100
* Visualisation/GPU nodes with AMD Radeon Pro WX8200, Nvidia Quadro RTX4000 or Nvidia A30
 
* Vector nodes with NEC Aurora TSUBASA CPUs
* Vector nodes with NEC Aurora TSUBASA CPUs
    
    
'''Features'''
'''Features'''
* Operating System: Centos 7
* Operating System: Rocky Linux 8
* Batchsystem: PBSPro
* Batchsystem: PBSPro
* node-node interconnect: Infiniband + GigE
* node-node interconnect: Infiniband + 10G Ethernet
* Global Disk 2.2 PB (lustre) for vulcan + 500TB (lustre) for vulcan2
* Global Disk 2.2 PB (lustre) for vulcan + 500TB (lustre) for vulcan2
* Many Software Packages for Development
* Many Software Packages for Development

Revision as of 20:42, 29 May 2024

Hardware

  • Pre- & Postprocessing node (smp node) [Entgeltordnung: 'smp Knoten']
    • 1 node
      • 2x Intel Xeon Gold 6148, 40 cores total @ 2.40GHz
      • 1.5TB memory
      • Nvidia Quadro K6000
      • 2TB HDD mounted on /tmp
      • 13TB HDD mounted on /localscratch
      • shared access
  • CascadeLake 40 cores compute nodes (clx)
    • 96 nodes (clx-25, clx384gb40c) [Entgeltordnung: 'Cascadelake 384 GB']
    • 8 nodes (clx-21, clx384gb40c-ai) [Entgeltordnung: 'CS500-Spark']
      • 2x Intel Xeon Gold 6230, 40 cores total @ 2.10GHz
      • 384GB memory
      • 1.8TB NVMe mounted at /localscratch
  • Haswell 24 Cores compute nodes (hsw) (only available on vulcan2)
    • 2x Intel Xeon E5-2668v3, 24 cores total @ 2.50GHz
    • 192 nodes (hsw128gb24c) [Entgeltordnung: 'Haswell 128 GB']
      • 128GB memory
  • Skylake 40 Cores compute nodes (skl) [Entgeltordnung: 'SKL-192GB']
    • 72 (skl192gb40c)
  • AMD 64 Cores compute nodes (genoa)
    • 60 nodes (genoa768gb64c)
      • 2x AMD Epyc 9334 Genoa, 2.7GHz base, 64 cores total
      • 768GB DDR5 memory
  • AMD 64 Cores SMP nodes (genoa-smp)
    • 2 nodes (genoa3072gb64c-smp)
      • 2x AMD Epyc 9334 Genoa, 2.7GHz base, 64 cores total
      • 3072GB DDR5 memory
  • Visualisation node with GPUs
    • 2x Intel Xeon Silver 4112, 8 cores total @ 2.60GHz
    • 96GB memory
    • 6 nodes (visamd) [Entgeltordnung: 'Graphikserver AMD']
      • AMD Radeon Pro WX8200
    • 1 node (visnv) [Entgeltordnung: 'Graphikserver NVIDIA']
      • Nvidia Quadro RTX 4000
  • Visualisation/GPGPU graphic nodes (visp100) [Entgeltordnung: 'VISP 100']
    • 10 nodes
      • 2x Intel Xeon E5-2667v4, 16 cores total @ 3.20GHz
      • 256GB memory
      • Nvidia Tesla P100 12GB
      • 3.7TB SSD mounted at /localscratch
      • 400GB SSD mounted at /tmp
  • AMD GPU nodes (genoa-a30)
    • 24 nodes
      • 2x AMD Epyc 9124 Genoa, 3.0GHz base, 32 cores total
      • 768GB DDR5 memory
      • 3.8TB NVMe mounted at /localscratch
      • Nvidia A30 with 24GB HBM2e
  • Interconnect: InfiniBand
    • HDR switches (200Gbit/s); 3 spine switches; each edge switch is connected to each spine switch with 4 cables
    • HDR100 connection to servers (100Gbit/s)
    • skl192gb40c nodes are connected to EDR switches (100Gbit/s)

Architecture

The NEC Cluster platform (vulcan) consists of several frontend nodes for interactive access (for access details see Access) and several compute nodes of different types for execution of parallel programs. Some parts of the compute nodes comes from the old NEC Cluster laki.

Compute node types installed:

  • Intel Xeon Broadwell, Skylake, CascadeLake
  • AMD Epyc Rome, Genoa
  • different Memory sizes (256GB, 384GB, 512GB, 768GB)
  • Pre-Postprocessing node with very large memory (1.5TB, 3TB)
  • Visualisation/GPU nodes with AMD Radeon Pro WX8200, Nvidia Quadro RTX4000 or Nvidia A30
  • Vector nodes with NEC Aurora TSUBASA CPUs

Features

  • Operating System: Rocky Linux 8
  • Batchsystem: PBSPro
  • node-node interconnect: Infiniband + 10G Ethernet
  • Global Disk 2.2 PB (lustre) for vulcan + 500TB (lustre) for vulcan2
  • Many Software Packages for Development