NIIF HPC (English)

Innen: KIFÜ Wiki

Magyar változat

Supercomputing

NIIF Institute has been operating supercomputer since 2001. The service provides scientific computation and data storage facilities. The components of the supercomputing infrastructure found in five locations:

  • NIIF centre Budapest
  • University of Debrecen
  • University of Pécs
  • University of Szeged
  • University of Miskolc


Az NIIF szuperszámítógépei térképen


Various supercomputer architectures are available in different locations, because we need to support different types of scientific computing tasks. In order to support all kinds of scientific computational needs different computer architectures are available. We have cluster based supercomputers in three locations and two ccNUMA machine. The supercomputers are operated by NIIF Institute staff. The latest development tools and scientific applications are available for the users. The documentation of the supercomputers can be found here.

Any person or research group who has a contractual relationship with the NIIF Institute can use the service. Usage of the supercomputers is free for authorised users. New users should request a HPC project here. After the project is accepted invitations can be sent out for group members by the principal investigator. Invitations for the HPC portal can be managed here.

NIIF HPC website

NIIF HPC resources

Location Budapest Budapest2 Szeged Debrecen Debrecen2-GPU (LEO) Debrecen3-Phi (Apollo) Pécs Miskolc
Type HP CP4000SL HP SL250s HP CP4000BL SGI ICE8400EX HP SL250s HP Apollo 8000 SGI UV 1000 SGI UV 2000
CPU / node 2 2 4 2 2 2 192 44
Core / CPU 12 10 12 6 8 12 6 8
Memory / node 66 GB 63 GB 132 GB 47 GB 125 GB 125 GB 6 TB 1.4 TB
Memory / core 2.6 GB 3 GB 2.6 GB 2.6 GB 7.5 GB 7.5 GB 5 GB 4 GB
CPU AMD Opteron 6174 @ 2.2GHz Intel Xeon E5-2680 v2 @ 2.80GHz AMD Opteron 6174 @ 2.2GHz Intel Xeon X5680 @ 3.33 GHz Intel Xeon E5-2650 v2 @ 2.60GHz Intel Xeon E5-2670 v3 @ 2.30GHz Intel Xeon X7542 @ 2.66 GHz Intel Xeon E5-4627 v2 @ 3.33 GHz
GPU - - 2 * 6 Nvidia M2070 - 68 * 3 Nvidia K20x + 16 * 3 Nvidia K40x - - -
Intel Xeon Phi (KNC) - 14 * 2 * Intel(R) Xeon Phi(TM) MIC SE10/7120 - - - 45 * 2 * Intel(R) Xeon Phi(TM) MIC SE10/7120 - -
Linpack Performance (Rmax) 5 Tflops 27 Tflops 20 Tflops 18 Tlops 254 Tflops ~106 Tflops 10 Tflops 8 Tflops
Compute nodes 32 14 50 128 84 45 1 1
Dedicated storage 50 TB 500 TB 250 TB 500 TB 585 TB (combined with Phi) 585 TB (combined with GPU) 500 TB 240 TB
Interconnect IB QDR IB NB FDR IB QDR IB QDR IB NB FDR IB NB FDR Numalink 6 Numalink 6
Scheduler SLURM SLURM SLURM SLURM SLURM SLURM SLURM SLURM
MPI OpenMPI (ompi) IntelMPI (impi) OpenMPI (ompi) SGI MPT (mpt) OpenMPI (ompi) OpenMPI (ompi) SGI MPT (mpt) SGI MPT (mpt)

Contact

If you have any questions send email to the following address: hpc-support@niif.hu

Or contact the HPC users in Hungary on the following mailing list: hpc-forum@listserv.niif.hu, you can subscribe here.

More information