Supercomputers/Systems/Lichtenberg I
Germany · decommissioned · hpc
Lichtenberg I
Also known as Lichtenberg-Hochleistungsrechner Phase I; TU Darmstadt HLR
Operated by TU Darmstadt .
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: 1263 of 1345 systems have a readable citation that names them, and 56 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
- 213.78 TFlop/s
- Rpeak
- 255.59 TFlop/s
- Rmax ÷ Rpeak
- 83.6%
- Cores
- 12,288
- Power
- 255.2 kW
- Per watt
- 0.8 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 | - IBM iDataPlex system | IBM | - | Reported | tu-darmstadt.detop500.org |
| Accelerator | NVIDIA Tesla K20X Accelerators CUDA (Kepler GK110) · 2,688C · 0.732 GHz · 28 nm · 235 W NVIDIA Tesla K20Xm GPUs, 43 nodes with 2 GPUs each | NVIDIA | 86 estimated | Reported | hkhlr.detu-darmstadt.de |
| Accelerator | - Intel Xeon Phi accelerator nodes, 20 nodes | Intel | 20 reported | Reported | tu-darmstadt.de |
| CPU | Intel Xeon E5-2670 CPUs x86-64 · 8C · 2.6 GHz · 32 nm · 115 W Intel Xeon E5-2670 8-core 2.6 GHz | Intel | - | Reported | top500.orghkhlr.de |
| Interconnect | Mellanox InfiniBand FDR Interconnect Fat tree · 56 Gb/s per 4x port · About 1 microsecond Mellanox InfiniBand FDR | NVIDIA | - | Reported | top500.org |
The read
The first Lichtenberg high-performance computer of TU Darmstadt and the Hessian HPC alliance, an IBM iDataPlex system inaugurated on 5 June 2013 with over 1,200 TFlop/s of combined computing power (23 percent from accelerators) at the time. The university describes over 700 standard nodes, four 1 TB large-memory nodes, 64 accelerator nodes (44 NVIDIA GPU, 20 Intel Xeon Phi), a 768 TB parallel file system, water cooling with free cooling and a 0.65 MW maximum draw once a second phase was added in 2014. The HKHLR cluster page lists the node type as dual Xeon E5-2670 (2 x 8 cores, 32 GB) with 43 nodes of two NVIDIA K20Xm GPUs, and states that Phase I was deactivated in April 2020. TOP500 lists the 12,288-core Xeon E5-2670 FDR InfiniBand part at 213.78 TFlop/s Rmax, 255.59 TFlop/s Rpeak and 255.2 kW, 148th in June 2013. Lichtenberg II is recorded separately.
Timeline
- 2013-06 List debut at #148 (June 2013) source
- 2013-06 First operational source
- 2020-04 Decommissioned source
Public list appearances
Pointers only: rank and edition, linking to the canonical entry. We do not reproduce list tables. See sourcing policy.
Change history
- 2026-10-09 Added Lichtenberg I
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 | does not name this system | Intel Xeon E5-2670, Mellanox InfiniBand FDR |
| tu-darmstadt.de | names system and part | Intel Xeon E5-2670 |
Checked 2026-10-13 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.