Supercomputers/Systems/AI-Farabium
Kazakhstan · operational · ai training
AI-Farabium
Also known as Al-Farabium; Kazakhtelecom GPU cloud
Operated by Kazakhtelecom .
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
- 17.93 PFlop/s
- Rpeak
- 26.84 PFlop/s
- Rmax ÷ Rpeak
- 66.8%
- Cores
- 57,600
- 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 | - Supermicro SYS-821GE-TNHR | Supermicro | - | Reported | top500.org |
| Accelerator | NVIDIA H200 SXM Accelerators CUDA (Hopper GH100) · 16,896C · 1.98 GHz · TSMC 4N · 700 W NVIDIA H200 SXM5 141GB, 400 GPUs per Kazakhtelecom | NVIDIA | 400 reported | Reported | top500.orgb2b.telecom.kz |
| CPU | Intel Xeon Platinum 8558 CPUs x86-64 · 48C · 2.1 GHz · Intel 7 Intel Xeon Platinum 8558 48C 2.1GHz | Intel | - | Reported | top500.org |
| Interconnect | NVIDIA InfiniBand NDR (Quantum-2) Interconnect Fat tree or dragonfly+ · 400 Gb/s per port · Under 0.6 microseconds ConnectX-7 400G Infiniband NDR | NVIDIA | - | Reported | top500.orgb2b.telecom.kz |
| Operating system | Ubuntu Server Operating systems Linux distribution · Open source, free with paid support tiers Ubuntu 22.04.4 LTS | Canonical | - | Reported | top500.org |
The read
Kazakhtelecom's NVIDIA H200 GPU cloud, listed on the TOP500 as AI-Farabium: a Supermicro SYS-821GE-TNHR cluster with Intel Xeon Platinum 8558 48C hosts, NVIDIA H200 SXM5 141GB GPUs, ConnectX-7 400G InfiniBand NDR and Ubuntu 22.04.4, 57,600 cores, Rmax 17.93 PFlop/s and Rpeak 26.84 PFlop/s at rank 103 on November 2025 and rank 122 on June 2026. Kazakhtelecom's B2B page describes 400 H200 GPUs with 56 TB of GPU memory, 100 TB of RAM, 6 PB of high-speed storage, 26.8 PFlop/s FP64 and 1,583 PFlop/s FP8, sold as virtual single-GPU servers or 8-GPU bare metal. This is a separate system from the Ministry-owned Alem.Cloud cluster in this dataset, although press coverage describes both as part of one platform. The data center location is not stated in the pages used.
Timeline
Public list appearances
Pointers only: rank and edition, linking to the canonical entry. We do not reproduce list tables. See sourcing policy.
Change history
- 2026-10-04 Added AI-Farabium, Kazakhtelecom's 400-GPU H200 cloud
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 |
|---|---|---|
| b2b.telecom.kz | names system, part and count | NVIDIA H200 SXM, NVIDIA InfiniBand NDR (Quantum-2) · counts: 400 |
| top500.org | names system, part and count | NVIDIA H200 SXM, Intel Xeon Platinum 8558, NVIDIA InfiniBand NDR (Quantum-2), Ubuntu Server · counts: 400 |
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.