Supercomputers/Systems/Supek
Croatia · operational · hpc
Supek
Operated by SRCE at SRCE Borongaj Data Centre .
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
- 1.25 PFlop/s
- Rmax ÷ Rpeak
- -
- Cores
- 8,384
- 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 | - Single HPE Cray EX2500 cabinet, fully integrated supercomputer | Hewlett Packard Enterprise | - | Reported | srce.unizg.hr |
| Accelerator | NVIDIA A100 SXM4 40GB Accelerators CUDA (Ampere GA100) · 6,912C · 1.41 GHz · 7 nm · 400 W 4 x NVIDIA A100 with 40 GB memory per GPU node | NVIDIA | - | Reported | srce.unizg.hr |
| CPU | AMD EPYC 7763 (Milan) CPUs x86-64 · 64C · 2.45 GHz · 7 nm · 280 W 2 x AMD Epyc 7763 2.45GHz per node | AMD | - | Reported | srce.unizg.hr |
| Interconnect | - Slingshot interconnection between nodes (generation not stated) | Hewlett Packard Enterprise | - | Reported | srce.unizg.hr |
| Storage | - HPE ClusterStor, 580 TB usable capacity, 238 GB/s write and 450 GB/s read | Hewlett Packard Enterprise | - | Reported | srce.unizg.hr |
The read
Croatia's first petascale supercomputer, operated by SRCE (the University of Zagreb's University Computing Centre) at the newly built data centre on University Campus Borongaj in Zagreb, commissioned in 2023 as part of the HR-ZOO (Croatian Scientific and Educational Cloud) project. SRCE's own page describes a single HPE Cray EX2500 cabinet with 2x AMD EPYC 7763 CPUs and 4x NVIDIA A100 40GB GPUs per GPU node (8,384 processor cores and 81 GPUs total), 32 TB of working memory, an HPE ClusterStor storage tier (580 TB usable, 238 GB/s write and 450 GB/s read) and a Slingshot interconnect between nodes, for a stated 1.25 PFLOPS peak. MEP d.o.o., the systems integrator that designed Supek's cooling and power infrastructure, independently corroborates the operator, the Borongaj campus location and the near-100kW single-rack cooling design, but does not restate the performance figure or exact hardware SKUs, so the row stays reported. No TOP500 submission was found for this system, so ranked is 0 and rmax_pflops is left blank with rpeak_pflops recorded as the operator-stated peak figure.
Timeline
- 2023 First operational source
Site & facility
SRCE Borongaj Data Centre
- Location
- Zagreb, Croatia
- Cooling
- Direct liquid cooling
- Commissioned
- 2023
Change history
- 2026-09-24 Supek added (Croatia)
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 |
|---|---|---|
| mep.hr | names system only | - |
| srce.unizg.hr | names system and part | AMD EPYC 7763 (Milan), NVIDIA A100 SXM4 40GB |
Checked 2026-10-09 by pnpm verify. Re-run it and the table changes with the web.
Further reading
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.
Operating a supercomputer
Acceptance testing, why "installed" and "in production" are six months apart, and the machine lifecycle.