NASA Discover

Also known as NCCS Discover

Operated by NASA at NASA Center for Climate Simulation (Goddard) .

Phase 1 seed dataset, compiled by hand. These rows were built from public operator, laboratory and vendor sources. A mechanical second-reader pass has since fetched every cited source: 1092 of 1153 systems have a readable citation that names them, and 40 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: 8.28 PFlop/s
Rmax ÷ Rpeak: -
Cores: 213,288
Power: 2.00 MW
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	- Supermicro TwinPro nodes (SCU16, SCU17, SCU18)	Supermicro SMCINASDAQ	-	Reported	nccs.nasa.gov
Accelerator	- NVIDIA A100 40GB, 4 per node on 12 GPU nodes in SCU16	NVIDIA NVDANASDAQ	48 reported	Reported	nccs.nasa.gov
CPU	- AMD EPYC Milan 64-core, 2 per node, 704 nodes each in SCU17 and SCU18	AMD AMDNASDAQ	-	Reported	nccs.nasa.gov
CPU	- Intel Xeon Cascade Lake Refresh 24-core, 2 per node, 676 nodes in SCU16	Intel INTCNASDAQ	-	Reported	nccs.nasa.gov
Interconnect	NVIDIA InfiniBand HDR Interconnect Fat tree or dragonfly+ · 200 Gb/s per port · About 0.6 microseconds InfiniBand HDR100 (100 Gbps) on CPU nodes, dual HDR 2x200 Gbps on GPU nodes	NVIDIA NVDANASDAQ	-	Reported	nccs.nasa.gov
Storage	- IBM GPFS file system, 49.5 PB disk	IBM IBMNYSE	-	Reported	nccs.nasa.gov
Operating system	SUSE Linux Enterprise Server Operating systems Linux distribution · Open source, commercial support subscription SUSE Linux Enterprise Server	SUSE	-	Reported	nccs.nasa.gov

The read

The NASA Center for Climate Simulation's general-purpose supercomputer at Goddard Space Flight Center, built as a Linux cluster from a 2006 base unit and grown by adding Scalable Compute Units, so the machine is an evolving heterogeneous cluster rather than a single procurement. NCCS lists 213,288 total cores at 8.28 PFlop/s aggregate performance, 49.5 PB of IBM GPFS disk and a 2 MW power draw. Its current SCU table lists SCU16 (676 Supermicro TwinPro nodes of Intel Xeon Cascade Lake Refresh, 1.9 PFlop/s, plus 12 Supermicro nodes with four NVIDIA A100 GPUs each), SCU17 and SCU18 (704 Supermicro TwinPro nodes each, 90,112 AMD EPYC Milan cores and 2.9 PFlop/s each) on HDR100 InfiniBand, with 11 older units retired between 2006 and 2024. NASA's SC23 exhibit says the system delivered 8.8 PFlop/s in 2023 and that NASA is planning a next-generation successor for hybrid HPC and AI workloads. NASA is the source for every figure and the totals reflect a system that changes as units are added, so the tier is reported.

Timeline

2006 First operational source

Site & facility

NASA Center for Climate Simulation (Goddard)

Location: Greenbelt, MD, USA
Commissioned: 2006

nccs.nasa.gov nas.nasa.gov

Change history

2026-10-01 Added NASA Discover (NASA)

Source check

We fetched this system's own citations and recorded whether each page actually mentions it. This is a corroboration signal, not a fact check, and it is published so you can see how well the sourcing holds up rather than take it on trust.

Cited source	Result	Found on the page
nccs.nasa.gov	names system only	-
nccs.nasa.gov	names system and part	NVIDIA InfiniBand HDR

Checked 2026-10-09 by pnpm verify. Re-run it and the table changes with the web.