DEGIMA
GPU cluster at Nagasaki University for astrophysical N-body simulations
DEGIMA is a high performance computing cluster run by Nagasaki University at the Nagasaki Advanced Computing Center. Built from 144 PCs linked by InfiniBand and packed with GeForce GPUs, it was designed for hierarchical N-body simulations and became known for its low hardware cost and energy use.
- 144 nodes with 144 CPUs and 576 GPUs in total
- Peak performance of 111 TFLOPS at 1376 MFLOPS per watt
- Hardware cost about US$500,000
- Runs N-body simulations with over 3,000,000,000 particles
- Named after the Dejima island in Nagasaki
What it is
DEGIMA, short for DEstination for Gpu Intensive MAchine, is a computing cluster operated by Nagasaki University at its Nagasaki Advanced Computing Center. It is a Beowulf-style cluster, meaning many ordinary PCs working together as one machine, and it is used for hierarchical N-body simulations. These track the gravitational interaction of huge numbers of particles and are a standard tool in astrophysics. The machine runs simulations with more than three billion particles using a method called the Multiple-Walk parallel treecode. In total it holds 144 processors and 576 graphics processors.
The compute
The cluster has 144 nodes joined by InfiniBand, a high speed network normally used to link machines inside a data centre. Every node is a PC built around an MSI X58 pro-E motherboard with a 2.66 GHz Intel Core i7 920 processor, 12 GB of DDR3-1333 memory and a Mellanox MHES14-XTC SDR InfiniBand adapter. Graphics work is handled by two GeForce GTX295 cards per node, and each card carries two GT200 GPU chips. Peak performance reaches 111 TFLOPS while drawing power at a rate of 1376 MFLOPS per watt.
Where the GPUs are
The system sits at the Nagasaki Advanced Computing Center within Nagasaki University in Nagasaki, Japan. Its name echoes Dejima, the small artificial island in Nagasaki harbour.
Money and ownership
The hardware for the whole machine cost roughly US$500,000. DEGIMA is noted for delivering its performance at low cost and with good energy efficiency.
Its place in AI
DEGIMA is used for astrophysical N-body work, simulating systems of over three billion particles with the Multiple-Walk parallel treecode. It is cited as an example of a cluster that achieves high throughput from consumer graphics cards rather than specialised supercomputing hardware.
Questions people ask about DEGIMA
What is DEGIMA used for?
It runs hierarchical N-body simulations, a technique in astrophysics for modelling how large numbers of particles attract one another gravitationally. The machine handles runs with more than three billion particles using the Multiple-Walk parallel treecode.
Who runs it and where is it?
Nagasaki University operates the cluster at the Nagasaki Advanced Computing Center in Nagasaki, Japan. The name comes from Dejima, a small artificial island in Nagasaki.
What hardware does it use?
Each of the 144 nodes is a PC with a 2.66 GHz Intel Core i7 920, 12 GB of memory and two GeForce GTX295 cards, each card holding two GT200 GPU chips. Nodes are linked over InfiniBand using Mellanox adapters.
How much did it cost and how efficient is it?
The hardware came to roughly US$500,000. The cluster reaches a peak of 111 TFLOPS and an energy efficiency of 1376 MFLOPS per watt, figures that have made it notable for cost and power savings.
Good reading and sources
The facts on this page come from the Wikipedia article "DEGIMA", 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 2026-11-14.