News
DealsCorvex agrees USD 103 million take-or-pay deal to perform AI inference with Mid-Atlantic GPUsSovereignNeysa scales AI infrastructure in India with AMD Instinct MI350X GPUs on TytanInferenceMicrosoft Foundry adds Baseten, DeepSeek AI inference (MSFT:NASDAQ)ClustersDataOne and Rowan University sign engineering co-op program agreement for DataOne's Vineland AI factoryClustersLook inside Cal Poly's $3 million AI Factory (Video)DealsOpenAI Taps AMD Chips for 457 MW of AI Capacity in Six-Year TensorWave DealClustersCal Poly launches an AI data center on campus. Take a look insideInferenceArista takes aim at the next AI network battleground: distributed inferenceDealsOpenAI Signs 457 MW AI Data Center Deal as AMD Expands Its Challenge to NvidiaClustersBehind Dell AI Factory’s NVIDIA & AMD Enterprise ExpansionDealsCorvex agrees USD 103 million take-or-pay deal to perform AI inference with Mid-Atlantic GPUsSovereignNeysa scales AI infrastructure in India with AMD Instinct MI350X GPUs on TytanInferenceMicrosoft Foundry adds Baseten, DeepSeek AI inference (MSFT:NASDAQ)ClustersDataOne and Rowan University sign engineering co-op program agreement for DataOne's Vineland AI factoryClustersLook inside Cal Poly's $3 million AI Factory (Video)DealsOpenAI Taps AMD Chips for 457 MW of AI Capacity in Six-Year TensorWave DealClustersCal Poly launches an AI data center on campus. Take a look insideInferenceArista takes aim at the next AI network battleground: distributed inferenceDealsOpenAI Signs 457 MW AI Data Center Deal as AMD Expands Its Challenge to NvidiaClustersBehind Dell AI Factory’s NVIDIA & AMD Enterprise Expansion
EveryNeocloud
Taiwania 3

Taiwania 3

A CPU-based supercomputer for scientific research and biomedical analysis in Taiwan.

Taiwania 3 is a high-performance computing cluster located at the National Center for High-Performance Computing in Taiwan. It utilises Intel processors to deliver approximately 2.7 petaflops of performance for public scientific research.

Share on WhatsApp

  • Uses Intel Xeon CPUs instead of GPUs for calculation.
  • Launched early in November 2020 for pandemic research.
  • Delivers 2.7 petaflops of peak computing power.
  • Operated by the Taiwanese government's NARLabs.
  • Supports biomedical and disaster management simulations.

What it is

Taiwania 3 is a large-scale computing system operated by the Taiwanese government through the National Center for High-Performance Computing. It serves as a resource for general scientific inquiry rather than artificial intelligence training. The machine relies entirely on central processing units from Intel to handle complex calculations. It provides open access to qualified scientists who seek approval from the operating centre. This system forms part of the broader Taiwania family of machines designed for national research needs.

How it came about

Construction began in 2019 under the direction of the National Center for High-Performance Computing. The team accelerated the timeline to meet urgent demands during the pandemic. The system officially launched in November 2020 ahead of schedule. It immediately joined efforts to study coronavirus data. By mid-2021, it was fully active and ranked on global performance lists. It replaced earlier generations while maintaining compatibility with existing software workflows.

The compute

The architecture centres on Intel Xeon Platinum 8280 processors running at 2.4 GHz. Each node contains two of these chips alongside substantial memory capacity. Interconnectivity relies on NVIDIA Mellanox InfiniBand technology for fast data transfer. The operating system is CentOS Linux managed by Slurm workload software. This setup prioritises traditional scientific computing over graphical processing tasks.

Where the GPUs are

The hardware resides within facilities managed by the National Center for High-Performance Computing. Manufacturing partners included Quanta Computer and ASUS Cloud for server assembly. The infrastructure supports specific environmental controls for temperature and humidity. Power consumption stands at roughly 564 kilowatts for the entire installation.

Customers and deals

Access is granted to academic and medical researchers after obtaining permission. The system supports biological laboratories connected to national databases. Public users can request time slots through service portals based on resource usage. Specific projects like film simulation algorithms also utilise its computational power.

Money and ownership

The project is funded by the state through NARLabs. No private venture capital or stock market listing applies to this entity. Costs are recovered through user fees for compute time. The total construction involved significant government investment in hardware procurement.

Its place in AI

This machine contributes to national disaster mapping and rescue planning. It helps combine sensor data into visual models for emergency response. Biomedical studies use it to analyse genetic sequences and drug interactions. It holds a position in global rankings for energy efficiency and raw speed.

Questions people ask about Taiwania 3

What type of chips does Taiwania 3 use?

It uses Intel Xeon Platinum 8280 central processing units. It does not rely on graphics cards for its primary computing tasks. This design suits general scientific research rather than deep learning.

Who can use this supercomputer?

Scientists and researchers can apply for access. Users must receive permission from the National Center for High-Performance Computing. Fees are charged based on the requested time and resources used.

Why was it built so quickly?

Construction started in 2019 but finished early in late 2020. The acceleration responded to urgent needs for COVID-19 research. It helped analyse viral data and support biomedical studies rapidly.

Good reading and sources

Where this page comes from

The facts on this page come from the Wikipedia article "Taiwania 3", written by its contributors and used under the CC BY-SA 4.0 licence, and from its Wikidata record (CC0). EveryNeocloud rewrote the text in its own words, with AI assistance, and compares every page with the article to make sure no passage is copied. It may be reused under the same CC BY-SA 4.0 licence with a link back.

It may contain mistakes, and figures change over time. Nothing here is investment or buying advice. If something is wrong, tell us. Last refreshed 2026-11-15.

Photo credits on this page

The small pictures above come from Wikimedia Commons under free licences. Some are cropped to fit.

The weekly digest

Every Monday: the provider of the day, new pages on EveryNeocloud, and the conferences and anniversaries of the week ahead.