Supercomputers/Systems/Mole-8.5
China · decommissioned · hpc
Mole-8.5
Also known as Mole-8.5 Cluster
Operated by IPE, Chinese Academy of Sciences .
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
- 496.50 TFlop/s
- Rpeak
- 1.01 PFlop/s
- Rmax ÷ Rpeak
- 49.0%
- Cores
- 29,440
- Power
- 540 kW
- Per watt
- 0.9 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 | - Mole-8.5 cluster built by IPE with Nvidia and Tyan, 288 server nodes | IPE, Chinese Academy of Sciences | - | Reported | top500.orgasianscientist.com |
| Accelerator | - NVIDIA Tesla 2050-class GPUs, more than 2,200 | NVIDIA | - | Verified | top500.orgasianscientist.com |
| CPU | - Intel Xeon X5520 4C 2.27GHz | Intel | - | Reported | top500.org |
| Interconnect | Mellanox InfiniBand QDR Interconnect Fat tree · 40 Gb/s per 4x port InfiniBand QDR | NVIDIA | - | Reported | top500.org |
The read
A GPU cluster of the Institute of Process Engineering, Chinese Academy of Sciences, built with Nvidia and Tyan: Xeon X5520 hosts with NVIDIA Tesla 2050-class GPUs on InfiniBand QDR. TOP500 lists 29,440 cores, 496.5 TFlop/s Rmax and 1,012.65 TFlop/s Rpeak at 540 kW, 43rd on the November 2012 list; it first appeared in November 2010 at rank 28 with 207 TFlop/s before the GPU upgrade. Asian Scientist independently names the same IPE cluster, 288 server nodes and more than 2,200 NVIDIA Tesla GPUs, used for the first whole-virus atomic simulation of H1N1 influenza. Status as of is the last TOP500 list it appeared on (June 2017); no retirement notice was found and the machine room location is not on any page fetched, so no site is recorded. TOP500 is the only source for Rmax, Rpeak and core count, so the tier is reported.
Timeline
- 2012-11 List debut at #43 (November 2012) 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
- 2026-10-02 Added Mole-8.5, the Chinese Academy of Sciences IPE GPU cluster
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 | names system and part | Mellanox InfiniBand QDR |
| asianscientist.com | names system only | - |
Checked 2026-10-09 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.