Taiwania
A series of supercomputers in Taiwan for scientific research and public access.
The Taiwania series comprises supercomputers owned by the National Applied Research Laboratories in Taiwan. The first system launched in May 2018 to support national research needs and has since expanded with higher-capacity models.
- Taiwania 1 launched on May 9, 2018, after a two-year build.
- Taiwania 2 reached 9 PFLOPS using NVIDIA V100 GPUs.
- Taiwania 3 was expedited in November 2020 for pandemic research.
- Systems are owned by the National Applied Research Laboratories.
- Taiwania 1 will be retired and replaced by Taiwania 4.
What it is
The Taiwania series represents a collection of high-performance computing systems based in Taiwan. These machines are owned by the National Applied Research Laboratories and operated through the National Center for High-performance Computing. The initiative began with Taiwania 1, which entered service in May 2018. Subsequent iterations, including Taiwania 2 and Taiwania 3, were developed to provide greater computational power for academic and scientific communities. The systems offer open access to researchers who obtain necessary permissions from the governing center.
How it came about
The project started with a two-year development phase costing at least NT$430 million. Taiwania 1 was officially activated on May 9, 2018. Taiwania 2 debuted later that year in November, ranking highly on performance lists. Taiwania 3 was accelerated into operation in November 2020 ahead of schedule to address demands related to the pandemic. In April 2023, plans were announced to retire Taiwania 1 and replace it with a new fourth iteration.
The compute
Hardware configurations vary across the series. Taiwania 1 features 3.4 petabytes of memory and achieves 1.33 quadrillion floating-point operations per second. Taiwania 2 utilizes 252 nodes, each equipped with dual Intel Xeon Gold CPUs and eight NVIDIA V100 GPUs, reaching 9 PFLOPS. Taiwania 3 employs Intel Xeon Platinum processors and InfiniBand HDR100 networking for 100 Gbit/s connectivity, delivering 2.7 PFLOPS. All systems run CentOS operating environments with Slurm workload management.
Where the GPUs are
The computing infrastructure is located within the National Center for High-performance Computing in Taiwan. This facility hosts the physical hardware for all known iterations of the Taiwania series. The center manages access rights and operational oversight for the machines, ensuring they serve the designated scientific community effectively.
Customers and deals
Access to the Taiwania supercomputers is primarily granted to scientists and researchers. Users must secure permission from the National Center for High-performance Computing before utilizing the resources. The systems are designated as open-access facilities intended to support specific research projects requiring significant computational capacity.
Money and ownership
The initial establishment of Taiwania 1 required an investment of at least NT$430 million. No further financial details regarding funding rounds or valuations are provided in the available records. The cost reflects the expenditure for the first generation of the series.
Its place in AI
Taiwania holds a position within Taiwan's broader supercomputing landscape alongside other regional systems like Taipei-1. It serves as a critical resource for local scientific advancement, particularly during periods of heightened demand such as the COVID-19 pandemic. The series demonstrates the country's commitment to maintaining advanced computational capabilities for public research.
Questions people ask about Taiwania
Who owns the Taiwania supercomputers?
The series is owned by the National Applied Research Laboratories in Taiwan. Operational control and access management are handled by the National Center for High-performance Computing.
What chips do the Taiwania systems use?
Taiwania 2 uses NVIDIA V100 GPUs paired with Intel Xeon Gold CPUs. Taiwania 3 utilizes Intel Xeon Platinum processors. Specific chip details for Taiwania 1 are not fully listed beyond its overall speed.
How can scientists access these computers?
The systems are open to the public for research purposes. Scientists must apply for and receive permission from the National Center for High-performance Computing to use the resources.
What is the current status of Taiwania 1?
As of April 2023, Taiwania 1 is scheduled for retirement. It will be replaced by a new system named Taiwania 4 to continue providing computational services.
Good reading and sources
- Taiwania 2 - QCT QuantaGrid D52G-4U/LC, Xeon Gold 6154 18C 3GHz, Mellanox InfiniBand EDR, NVIDIA Tesla V100 SXM2TOP500, Article
- Supercomputing ArchivesNvidia, Article
- Taiwania (supercomputer) on WikipediaEncyclopedia
- More photos on Wikimedia CommonsPhotos
- Wikidata recordOpen data
The facts on this page come from the Wikipedia article "Taiwania (supercomputer)", 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.
Like Taiwania
All AI supercomputers
🇨🇭 SwitzerlandAlps
Swiss national supercomputer using Nvidia GH200 chips for AI and scientific research.Supercomputer · Other Nvidia GPUs · 2024
🇺🇸 United StatesNCAR-Wyoming Supercomputing Center
A high-performance computing and data archival facility in Cheyenne, Wyoming.Supercomputer · Other Nvidia GPUs · 2003
🇩🇪 Germany and 4 moreNorthern Data
German GPU cloud and data centre group for AI and high-performance computingNeocloud · Nvidia Hopper · 2019
🇷🇺 RussiaChristofari
Sberbank's Nvidia-based supercomputers in Moscow for neural network trainingSupercomputer · Other Nvidia GPUs · 2019
More from National Applied Research Laboratories
All of them
🇯🇵 JapanFugaku
A Japanese CPU-based supercomputer built by Fujitsu for RIKEN research work.Supercomputer · Other and custom chips · 2020
🇵🇹 PortugalDeucalion
Portugal's fastest supercomputer, hosted at MACC in GuimarãesSupercomputer · Nvidia Ampere · 2023