Supercomputers/Systems/Edge (LLNL)
USA · operational · hpc
Edge (LLNL)
Also known as Edge Appro GreenBlade
Operated by Lawrence Livermore National Laboratory at Livermore Computing Center .
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
- 100.50 TFlop/s
- Rpeak
- 239.87 TFlop/s
- Rmax ÷ Rpeak
- 41.9%
- Cores
- 8,240
- Power
- 745 kW
- Per watt
- 0.1 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 Edge (LLNL), and those in USA. 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 | 100.50 TFlop/s | #118 of 177 · bottom 35% | #249 of 332 · bottom 26% |
| Energy efficiency | 0.1 GF/W | #116 of 129 · bottom 11% | #155 of 162 · bottom 5% |
| Rmax ÷ Rpeak | 41.9 % | #177 of 177 · bottom 1% | #326 of 331 · bottom 3% |
| Power draw | 745 kW | #44 of 131 · top 33% | #76 of 171 · top 44% |
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 | - Appro GreenBlade cluster | Appro HPENYSE
rolls up to Hewlett Packard Enterprise | - | Reported | top500.orgweb.archive.org |
| Accelerator | NVIDIA Tesla M2050 Accelerators CUDA (Fermi) · 448C · 1.15 GHz · 40 nm · 225 W NVIDIA Tesla M2050, two per compute node | NVIDIA | 412 reported | Reported | web.archive.orgtop500.org |
| CPU | - Intel Xeon X5660 6C 2.8 GHz, two per node, 2,592 CPU cores | Intel | - | Reported | web.archive.orgtop500.org |
| Interconnect | Mellanox InfiniBand QDR Interconnect Fat tree · 40 Gb/s per 4x port InfiniBand QDR (Mellanox) | Mellanox NVDANASDAQ
rolls up to NVIDIA | - | Reported | web.archive.orgtop500.org |
The read
Edge was Livermore's unclassified visualisation and data-analysis cluster, described on LLNL's archived 2011 and 2013 pages as the only production resource at Livermore Computing with GPUs and as limited-access. Those pages list 216 nodes (2 login, 206 batch) with two six-core Xeon X5660 2.8 GHz CPUs each (2,592 CPU cores), 412 Tesla M2050 GPUs (two per compute node), 48 or 96 GB per node, InfiniBand QDR from Mellanox and fast links to the Lustre file systems. TOP500 lists the Appro GreenBlade cluster at 8,240 cores (GPU cores evidently counted), 100.5 TFlop/s Rmax, 239.87 TFlop/s Rpeak and 745 kW across six lists, 73rd in November 2010 to 470th in June 2013.
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
Livermore Computing Center
- Location
- Livermore, CA, USA
- Cooling
- Direct liquid cooling (warm water)
- Commissioned
- 1996
Change history
- 2026-10-18 Edge (LLNL) added from its source pages
Nearby in scale
The machines closest to Edge (LLNL) by Rmax, smaller to larger.
Related systems
Same site
- Hera (LLNL) · USA, operational, 2009
- Zin · USA, decommissioned, 2011
- Cab · USA, decommissioned, 2012
- Sequoia · USA, decommissioned, 2012
- Vulcan · USA, decommissioned, 2012
- Atlas (LLNL) · USA, decommissioned, 2007
Same operator
- ASC Purple · USA, decommissioned, 2005
- BlueGene/L · USA, decommissioned, 2004
- Jade (LLNL) · USA, operational, 2016
- Quartz (LLNL) · USA, decommissioned, 2016
- Thunder · USA, decommissioned, 2004
- Ansel (LLNL) · USA, operational, 2018
Same accelerator
- Tianhe-1A · China, operational, 2010
- TSUBAME2.0 · Japan, upgraded, 2010
- HokieSpeed · USA, decommissioned, 2011
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.