Supercomputers/Systems/CRESCO6
Italy · operational · hpc
CRESCO6
Also known as ENEA CRESCO6
Operated by ENEA at ENEA Portici Research Centre .
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 1581 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
- 1.01 PFlop/s
- Rpeak
- 1.40 PFlop/s
- Rmax ÷ Rpeak
- 72.1%
- Cores
- 20,832
- 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.
How it compares
Ranked against the other machines in this dataset that report the same figure: those that first came into service within two years of CRESCO6, and those in Italy. Computed from our tables; a metric is ranked only when at least eight peers report it.
| Metric | Value | Against its era | Against its country |
|---|---|---|---|
| Rmax | 1.01 PFlop/s | #175 of 223 · bottom 23% | #18 of 21 · bottom 16% |
| Rmax ÷ Rpeak | 72.1 % | #84 of 216 · top 39% | #9 of 20 · top 42% |
Every metric is ranked largest-first, so a high power-draw rank means a big power draw, not a good one. Rmax ÷ Rpeak uses HPL-measured results only. Rankings move as machines are added; they describe this dataset, not the world.
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 | - Lenovo SD530 | Lenovo | - | Reported | top500.org |
| CPU | Intel Xeon Platinum 8160 CPUs x86-64 · 24C · 2.1 GHz · 14 nm · 150 W Intel Xeon Platinum 8160 24C 2.1GHz, two per node (216 nodes in the first stage) | Intel | - | Verified | top500.orgict.enea.it |
| Interconnect | Omni-Path Architecture Interconnect Fat tree · 100 Gb/s per port · About 1 microsecond Intel Omni-Path 100 Series, 48-port edge switches, fat tree | Cornelis Networks | - | Verified | top500.orgict.enea.it |
The read
ENEA's Skylake and Omni-Path Tier-1 cluster at Portici, bought after the 2015 ENEA-CINECA agreement to serve EUROfusion and predecessor to CRESCO8 (enea-cresco8). ENEA's own poster describes the first stage: 216 nodes, 10,368 cores, two Xeon Platinum 8160 per node with 192 GB, a 15-switch Omni-Path fat tree, about 0.7 PFlop/s nominal, 434 TFlop/s HPL and 95 kW during HPL. TOP500 ranked it 420th in November 2018 with 20,832 cores, which is 434 nodes of 48 cores, so the machine had roughly doubled by then (inferred from the core count, not stated on the poster); Rmax 1.01 and Rpeak 1.40 PFlop/s are TOP500 figures and the power field is blank there. TOP500 lists Lenovo as the manufacturer (SD530 nodes) with SUSE Linux. An ENEA notice of December 2021 says it was reopened for only half its nodes, and a November 2024 Punto Informatico article still describes CRESCO6 and CRESCO7 as active at Portici beside the new CRESCO8.
Timeline
Public list appearances
Pointers only: rank and edition, linking to the canonical entry. We do not reproduce list tables. See sourcing policy.
Site & facility
Change history
- 2026-10-29 Added CRESCO6, ENEA's Skylake and Omni-Path cluster at Portici
Nearby in scale
The machines closest to CRESCO6 by Rmax, smaller to larger.
- ASPIRE 1 · Singapore, decommissioned · 1.01 PFlop/s
- Zenith (Dell EMC) · USA, operational · 1.01 PFlop/s
- Lemhi · USA, decommissioned · 1 PFlop/s
- Tryton · Poland, operational · 1.01 PFlop/s
- PSNC Eagle · Poland, operational · 1.01 PFlop/s
- MareNostrum P9 CTE · Spain, operational · 1.02 PFlop/s
Related systems
Same site
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.