Supercomputers/Systems/Olivia
Norway · operational · mixed
Olivia
Operated by Sigma2 at Lefdal Mine Data Center .
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
- 13.20 PFlop/s
- Rpeak
- 16.80 PFlop/s
- Rmax ÷ Rpeak
- 78.6%
- Cores
- -
- Power
- 219 kW
- Per watt
- 60.3 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 | - HPE Cray Supercomputing EX (EX4000 platform) | Hewlett Packard Enterprise | - | Reported | sigma2.notop500.org |
| Accelerator | NVIDIA GH200 Grace Hopper Superchip Accelerators Arm CPU + CUDA GPU · TSMC 4N 448 NVIDIA GH200 Grace Hopper superchips (304 at launch, 144 added February 2026) | NVIDIA | 448 reported | Reported | sigma2.nodocumentation.sigma2.no |
| CPU | AMD EPYC 9005 (Turin) CPUs x86-64 · 192C · 2.7 GHz · TSMC N3E · 500 W 252 CPU nodes, two 128-core AMD EPYC Turin (Zen 5c) CPUs each, 64,512 cores | AMD | - | Reported | documentation.sigma2.no |
| Interconnect | HPE Slingshot-11 Interconnect Dragonfly · 200 Gb/s per port · Around 2 microseconds HPE Slingshot, Dragonfly topology, 200 Gbit/s | Hewlett Packard Enterprise | - | Reported | documentation.sigma2.no |
| Storage | HPE Cray ClusterStor E1000 Storage Lustre appliance, flash and disk tiers · 700 PB on Frontier (Orion) · About 10 TB/s peak HPE Cray ClusterStor E1000, 5.3 PB | Hewlett Packard Enterprise | - | Reported | sigma2.no |
| Cooling | - Direct liquid cooling of compute and accelerator nodes | Hewlett Packard Enterprise | - | Reported | documentation.sigma2.no |
The read
Norway's national flagship system, an HPE Cray EX at the Lefdal Mine Data Center in Nordfjordeid, procured by Sigma2 under a contract signed in June 2024 (225 million NOK) and inaugurated on 17 June 2025 with full service in autumn 2025. Sigma2's documentation lists 252 CPU nodes with two 128-core AMD EPYC Turin CPUs each (64,512 cores) and 76 accelerator nodes carrying 304 NVIDIA GH200 superchips on HPE Slingshot with a Dragonfly topology. Sigma2 announced on 27 February 2026 that 144 more GH200 superchips were added, for 448 in total. The 13.20 PFlop/s Rmax, 16.80 PFlop/s Rpeak and 219 kW figures are TOP500's for the original 304-superchip configuration (rank 134 on the November 2025 list); the core count is left blank because the expansion post-dates those runs and no post-expansion figure was confirmed.
Timeline
- 2024-06 Announced source
- 2025-06 First operational source
- 2025-11 List debut at #134 (November 2025) source
Public list appearances
Pointers only: rank and edition, linking to the canonical entry. We do not reproduce list tables. See sourcing policy.
Site & facility
Lefdal Mine Data Center
- Location
- Nordfjordeid, Norway
- Cooling
- Direct liquid cooling
- Commissioned
- 2025
Change history
- 2026-10-01 Added Olivia (Sigma2)
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 |
|---|---|---|
| top500.org | names system and part | NVIDIA GH200 Grace Hopper Superchip, HPE Slingshot-11, HPE Cray ClusterStor E1000 |
| sigma2.no | names system, part and count | NVIDIA GH200 Grace Hopper Superchip, HPE Cray ClusterStor E1000 · counts: 448 |
Checked 2026-10-09 by pnpm verify. Re-run it and the table changes with the web.
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.