About PARAM Shivay
The supercomputer of 838 TFLOPS capacity, built at the cost of Rs 32.5 crore under the National Super Computing Mission at Indian Institute of Technology (IIT), Banaras Hindu University (BHU).
PARAM Shivay Details
| System Specifications | |
|---|---|
| Theoretical Peak Floating-point Performance Total (Rpeak) | 837 TFLOPS |
| Base Specifications (Compute Nodes) | 2 X Intel Xeon Skylake 6148, 20 Cores, 2.4 GHz, Processors per node, 192 GB Memory, 480 GB SSD |
| Master/Service/Login Nodes | 10 nos. |
| CPU only Compute Nodes (Memory) | 192 nos. (192GB) |
| GPU Nodes (Memory) | 11 nos. (22 Nvidia V100 PCIe) |
| High Memory Compute Nodes | 20 nos. (768GB) |
| Total Memory | 54.3 TB |
| Interconnect | Primary: 100Gbps Mellanox Infiniband Interconnect network 100% non-blocking, fat tree topology Secondary: 10G/1G Ethernet Network Management network: 1G Ethernet |
| Primary Storage | Lustre based Primary storage 750 TiB usable with 25 GB/Sec write throughput |
| Archival Storage | Archival Storage 250 TiB usable capacity based on DDN Gridscaler (GPFS) with 1GB/sec write throughput |
| CPU Only Compute Nodes | |
|---|---|
| Nodes | 192 |
| Cores | 7680 |
| Compute Power of Rpeak | 589 TFLOPS |
| Each Node with | 2 X Intel Xeon Skylake 6148, 20 cores, 2.4 GHz, processors 192 GB memory 480 GB SSD |
| GPU Only Compute Nodes | |
|---|---|
| Nodes | 11 |
| CPU Cores | 440 |
| CUDA Cores | 112640 |
| Rpeak | Compute Rpeak 187 TF |
| Each Node with | 2 X Intel Xeon Skylake 6148, 20 cores, 2.4 GHz, processors 192 GB Memory 2 x NVIDIA V100 PCIe Accelerator Cards 16 GB HBM2 480 GB SSD |
| High Memory Compute Nodes | |
|---|---|
| Nodes | 20 |
| CPU Cores | 800 |
| Compute Power of Rpeak | 61 TFLOPS |
| Each Node with | 2 X Intel Xeon Skylake 6148, 20 cores, 2.4 GHz, processors 768 GB Memory 480 GB SSD |
Architecture Diagram:
Software Stack:
Installed Applications/Libraries
HPC Applications
- Bio-informatics: MUMmer, HMMER, MEME, PHYLIP, mpiBLAST, ClustalW
- Molecular Dynamics: NAMD (for CPU and GPU), LAMMPS, GROMACS
- CFD: OpenFOAM, SU2
- Material Modeling, Quantum Chemistry: Quantum-Espresso, Abinit, CP2K, NWChem
- Weather, Ocean, Climate: WRF-ARW, WPS (WRF), ARWPost (WRF), RegCM, MOM, ROMS
Deep Learning Libraries
- cuDNN, TensorFlow, Theano
Dependency Libraries
- NetCDF, PNETCDF, Jasper, HDF5, Tcl, Boost, FFTW
Support
For any support, contact: shivaysupport@iitbhu.ac.in
PARAM Shivay Usage Report
Note: The above data is coming from C-Chakshu (Multi Cluster Monitoring Platform)
Publication
There are a total of 219 publications that have been published using PARAM Shivay till 2023.