Skip to main content

Benchmark Center

When deciding on HPC purchases, managers are often faced with the question of which system is the best system for their individual application. In addition to performance, the overall cost and energy requirement of the chosen solution are also crucial factors.

Practical performance measurement is a key aspect when it comes to identifying potential for optimization. Benchmark analyses help you select the right HPC components for you. We therefore work closely with the sector’s technology leaders and offer the option of testing your applications using up-to-date hardware not yet available on the market.

Thanks to its many years’ experience, the MEGWARE Benchmark Center boosts the necessary methodical expertise for the meaningful comparison of figures.Our HPC and AI engineers will be more than happy to advise you on the best configuration for you and provide you with access to the following hardware and software remotely or on site at our company:

Available Hardware

  • HPC-Server von MEGWARE, Intel, Supermicro, Quanta, Gigabyte und ASUS
  • Prozessoren: Intel Xeon Scalable (Cascade Lake, Cascade Lake AP, Ice Lake), AMD Epyc (Rome, Milan, Milan-X), Huawei ARM v8.2 Kunpen 920
  • Beschleuniger: NVIDIA A100 & H100, AMD Instinct (MI200 Series) und Xilinx Alveo U280
  • Parallele Filesysteme von Panasas, IBM Spectrum Scale (GPFS), Lustre und BeeGFS mit einer Bandbreite von mindestens 1GB/s
  • Infiniband Interconnect EDR oder HDR von NVIDIA und OmniPath von Cornelis Networks
  • DDR4 Speicher von Samsung und Micron
Gigabyte R181-Z901AMD Epyc (Naples/Rome)up to 512GB DDR4-3200HDDs, SSDs, PCIe-NVMEup to 2
Gigabyte H261-Z601 system with 4 nodesAMD Epyc (Naples/Rome)up to 512GB DDR4-3200 per nodeHDDs, SSDs, PCIe-NVMEn/a
Gigabyte H262-Z611 system with 4 nodesAMD Epyc (Rome) with PCI Gen4up to 512GB DDR4-3200 per nodeHDDs, SSDs, PCIe-NVMEn/a
Intel S2600BP12 systems with 4 nodesIntel Scalable (Skylake / Cascade Lake)up to 384GB DDR4-2933 per nodeHDDs, SSDs, PCIe-NVMEn/a
Intel S2600WF2Intel Scalable (Skylake / Cascade Lake)up to 786GB DDR4-2933HDDs, SSDs, PCIe-NVMEup to 2
TaiShan 200 22804Kunpeng 920up to 512GB DDR4-2933SSDs, NVMEsn/a
ManufacturerTypeCoresClock in GHzQuantity
AMDEpyc 7F52163,52
AMDEpyc 7452322,356
AMDEpyc 7402242,82
AMDEpyc "Milan" 7713642,08
AMDEpyc "Milan-X" 7773x642,28
IntelXeon Scalable 6252242,14
IntelXeon Scalable 6238R382,22
IntelXeon Scalable 6240Y182,68
IntelXeon Scalable 6258R282,72
IntelXeon Scalable 8362322,82
IntelXeon Scalable 8368382,42
IntelXeon "IceLake" 6346163,12
IntelXeon "IceLake" 8360Y362,48
HuaweiKunpeng 920-4826482,68
AMDInstinct MI2102
Xilinx AlveoU280 (PCIe Gen4)1
IntelOptane DC P4800X375GBPCIe-NVME2
IntelOptane DC P4800X375GBU.28
IntelDC P45104TBU.24
IntelDC S350080GBSATA4
Micron1100 SED256GBSATA1
Western DigitalDC SN6401.6TBU.22
Cornelis NetworksOmniPath100GBitPCIe3 x1612
NVIDIAConnect-X6 HDR InfiniBand200GBitPCIe4 x164
NVIDIAConnect-X4 EDR InfiniBand100GBitPCIe3 x164

Available Software

  • Operating systems: Linux (Red Hat, SUSE) and Windows HPC Server
  • Compilers: Intel C and Fortran compilers, GCC
  • MPI implementations : Intel MPI, Open MPI, MVAPICH
  • Optimized software libraries, such as ACML and Intel MKL
  • Cluster management software ClustWare® from MEGWARE
  • Intel Parallel Studio Cluster Edition including Intel® VTune™ Amplifier XE, Intel® Inspector and Intel® Advisor
  • Artificial Intelligence  / Machine Learning Frameworks: caffe2. cntk, digits, mxnet, pytorch, tensorflow tensorrt, theano, torch

We will be more than happy to perform a proximate analysis for you, free of charge. Further benchmarks and optimizations will be charged for at cost.

Tobias Pfennig

Tobias Pfennig
Sales Director