Supercomputers/Systems/PARAM Rudra (S. N. Bose)
India · operational · hpc
PARAM Rudra (S. N. Bose)
Operated by SNBNCBS .
Researched with AI assistance, checked by scripts. These rows were researched by AI agents reading public operator, laboratory and vendor sources, appended by scripts and checked by automated validation; no subject-matter expert has reviewed them (how this site is made). A mechanical second-reader pass has since fetched every cited source: 1350 of 1540 systems have a readable citation that names them, and 64 are genuinely weakly sourced. Every claim carries its source and a confidence tier. Treat anything below verified as a lead, not a citation.
Measured performance
- Rmax
- -
- Rpeak
- 838.00 TFlop/s
- Rmax ÷ Rpeak
- -
- Cores
- -
- Power
- -
- Per watt
- - GF/W
Figures are as last publicly reported for the configuration described below, not a live measurement. Where a system was upgraded in place, the post-upgrade configuration is shown and the earlier one appears in the timeline.
What this machine is made of
One row per supplier relationship. “Supplier at build” is the company that shipped the part at the time; where that company has since been acquired, the parent it rolls up to today is shown beside it. That distinction is what makes ticker-level aggregation possible across a thirty-year dataset.
| Role | Part | Supplier at build | Quantity | Confidence | Source |
|---|---|---|---|---|---|
| Integrator | - C-DAC designed and commissioned; Rudra servers supplied by VVDN Technologies | C-DAC | - | Reported | nsmindia.incdac.in |
| Accelerator | - 8 GPU-only compute nodes with 2 x NVIDIA A100 each | NVIDIA | 16 reported | Reported | nsmindia.in |
| CPU | - 2 x Intel Xeon Gold 6240R 24C 2.4GHz per node | Intel | - | Reported | nsmindia.in |
| Interconnect | - 100Gbps Mellanox InfiniBand, 100% non-blocking fat tree | NVIDIA | - | Reported | nsmindia.in |
The read
The 838 TFlop/s PARAM Rudra at the S. N. Bose National Centre for Basic Sciences in Kolkata, dedicated on 27 September 2024 with the IUAC (param-rudra-iuac) and GMRT (param-rudra-gmrt) machines. The NSM sheet lists 96 CPU-only nodes, 32 high-memory (768 GB) nodes, 26 GPU-ready nodes and 8 GPU nodes with two A100 cards each (16 GPUs), all on dual Xeon Gold 6240R, with 49.536 TB of memory, a 100 Gb/s Mellanox InfiniBand non-blocking fat tree and 1.0 PiB of storage. C-DAC's release states the 838 TFlop/s figure and that the Rudra servers in these three facilities came from VVDN Technologies. No Linpack result is published.
Timeline
- 2024-09 First operational source
Change history
- 2026-10-18 Added PARAM Rudra (S. N. Bose) (SNBNCBS)
Further reading
Operating a supercomputer
Acceptance testing, why "installed" and "in production" are six months apart, and the machine lifecycle.
Inside a node
Latency-optimised versus throughput-optimised processors, NUMA, and the end of the host-to-device copy.
The interconnect
Topologies, why latency and tail behaviour matter more than bandwidth, and how the fabric market consolidated.