Supercomputers/Systems/Pengcheng Cloudbrain-II
China · operational · ai training
Pengcheng Cloudbrain-II
Operated by Peng Cheng Laboratory at Peng Cheng Laboratory (Shibilong, Shenzhen) .
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
- -
- Rpeak
- -
- Rmax ÷ Rpeak
- -
- Cores
- -
- 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.
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 | - Huawei Atlas 900 AI cluster | Huawei | - | Verified | prnewswire.comeyeshenzhen.com |
| Accelerator | Huawei Ascend 910 Accelerators Da Vinci · 7 nm · 310 W thousands of Ascend 910 AI processors | Huawei | - | Reported | prnewswire.com |
| Interconnect | Huawei HCCS Interconnect Mesh within a node HCCS, PCIe 4.0 and 100G RoCE | Huawei | - | Reported | prnewswire.com |
| Operating system | Kylin Linux Operating systems Linux distribution · State-endorsed Chinese distribution KylinOS v10sp1 on the IO500 client nodes | KylinSoft | - | Reported | io500.org |
The read
Huawei and Peng Cheng Laboratory's AI platform in Shenzhen, built on Huawei Atlas 900 clusters of Ascend 910 accelerators. Huawei announced it on 29 November 2019 at 100 PFlop/s with a target of 1,000. Its only published figure is 256 to 1,024 PFlop/s of FP16 for the Atlas 900 product, which is an AI rate and not a measured HPL result, so no Rmax or Rpeak is recorded. It has never appeared on TOP500 and is known through IO500 storage results, where its file system ran on 300 client nodes. When it started work is disputed between the two sources, and the count of Ascend chips is not published, so no quantity is recorded.
Timeline
- 2019-11 Announced source
Site & facility
Where sources disagree
We record conflicts instead of picking a winner quietly. All open questions.
- Start of operation: Phase one launched 29 November 2019 at 100 PFlop/s (Huawei release) against Settled in Shenzhen in September 2020 with tests from October 2020 (Shenzhen news). No first-operational date is recorded.
Change history
- 2026-09-20 Added as an unranked AI platform with no measured HPL figure
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 |
|---|---|---|
| eyeshenzhen.com | names system and part | Huawei Ascend 910, Huawei HCCS |
| prnewswire.com | does not name this system | Huawei Ascend 910, Huawei HCCS |
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
Why accelerators exist, what the memory hierarchy costs, and how the CPU and accelerator merged onto one package.
The interconnect
Topologies, why latency and tail behaviour matter more than bandwidth, and how the fabric market consolidated.