DAEDALUS node architecture
DAEDALUS is operated by GRNET · Lavrion Technological and Cultural Park, NTUA in Greece. This profile covers Ranked system / partition. Specifications are listed below; anything the operator has not published is itemised in the documented-fields list.
Published benchmark (TOP500 · June 2026)
- HPL rank
- #31
- HPL Rmax
- 85.69 PFlop/s
- Theoretical peak Rpeak
- 115.08 PFlop/s
- Ranked core count
- 411,264
- Measured power
- 1395 kW
- Energy efficiency
- 61.43 GFlop/s per watt
Cluster · GH200 compute
- Operator / site
- GRNET · Lavrion Technological and Cultural Park, NTUA
- Country
- Greece
- Profile scope
- Ranked system / partition
- Data checked
- 14 September 2026
- Manufacturer
- HPE
- Ranked core count
- 411264
- Theoretical peak (PFlop/s)
- 115.08
- Measured power (kW)
- 1395
- Energy efficiency (GFlop/s per watt)
- 61.42652329749104
- Operating system
- Red Hat Enterprise Linux
GH200 compute
Items marked ▧ come from the documented NVIDIA GH200 Grace Hopper Superchip platform specification.
- System / node model
- HPE Cray EX254n
- CPU model
- NVIDIA Grace · 72 Arm Neoverse V2 cores
- CPU note
- DAEDALUS is partitioned by core: the CPU partition uses 64 of each GH200 superchip’s 72 Arm cores, while the accelerated partition uses the remaining 8 cores together with the H100.
- Accelerator model
- NVIDIA GH200 Superchip
- CPU ↔ accelerator link
- NVLink-C2C · 900 GB/s coherent CPU ↔ GPU
Scope and uncertainty
GRNET publishes no DAEDALUS user documentation: daedalus.grnet.gr resolves but is not served, and the doc and docs subdomains do not answer. Its 2023 public-consultation tender annex specifies separate CPU-only and accelerated node types, which the delivered machine does not use, so none of its figures is carried over. DAEDALUS presents two virtual partitions over the same GH200 hardware: a CPU partition using 64 of the 72 Arm cores per superchip, and a GPU partition using the remaining 8 cores plus the H100. GRNET publishes no node count and no per-node configuration. The 2022 tender document states requirements rather than the as-built machine and is not used as a source. Node specifications follow the documented NVIDIA GH200 Grace Hopper Superchip platform (Superchip module — nodes per system and per-node superchip counts are operator-specific); operator-specific deviations are not published.
Documented fields: 11 / 27
Every field this catalog checks. Fields listed above have a value; the remainder are shown here. A field marked absent is not publicly available in the sources we could verify, which does not mean the hardware is missing. Fields that cannot apply, such as benchmark results for an unranked system, are excluded from the total.
Cluster — 6 / 7
- ✓Manufacturer
- ✓Ranked core count
- ✓Theoretical peak (PFlop/s)
- ✓Measured power (kW)
- ✓Energy efficiency (GFlop/s per watt)
- ✓Operating system
- ✕Network topology
GH200 compute — 5 / 20
- ✓System / node model
- ✕Compute nodes of this type
- ✓CPU model
- ✕CPU sockets per node
- ✕Cores per socket
- ✕Physical CPU cores
- ✓CPU note
- ✓Accelerator model
- ✕Physical accelerators per node
- ✕Host / unified memory
- ✕Accelerator memory
- –Accelerator ↔ accelerator link
- –On-node fabric
- ✓CPU ↔ accelerator link
- ✕Network interfaces per node
- ✕Interface model
- ✕Interface speed
- ✕Interface-to-accelerator topology
- –Aggregate scale-out per node
- ✕GPUDirect RDMA
- ✕Management / storage networking
- ✕Local storage
- –NVMe data disks
- –Max remote data disks
- –Uncached disk IOPS
- –Uncached disk throughput (MBps)
- ✕Form factor
Sources
Browse all cluster architecture pages →
CLUSTERATLAS