The 29th edition of the TOP 500 supercomputers awarded the top spot to the BlueGene/L System developed by IBM for the US Department of Energy's National Nuclear Security Administration.
The machine recorded a Linpack benchmark performance of 280.6 TFlop/s (trillions of calculations per second).
Second place was taken by an upgraded Jaguar-Cray XT4/XT3 housed at the Oak Ridge National Laboratory with a benchmark performance of 101.7 TFlop/s.
The only other system to exceeded 100 TFlop/s was Sandia National Laboratory's Cray Red Storm system which ranked third at 101.4 TFlop/s.
IBM and HP continue to dominate the market, having sold the bulk of the systems at all performance levels.
But the current list shows that HP has edged out IBM with 40.6 percent of the total number of systems on the list, compared with IBM's 38.4 percent.
No other manufacturer has more than five per cent of the systems on the list, but Dell came close at 4.8 percent.
However, IBM remains the clear leader in performance with 41.9 percent of installed performance compared to HP with 24.5 percent and Dell with a surprising 8.8 percent.
Two new IBM BlueGene/L systems entered the Top 10, at the New York Centre for Computational Science in Stony Brook, New York (ranked fifth) and the Computational Center for Nanotechnology Innovations at Rensselaer Polytechnic Institute in Troy, New York (ranked seventh).
These two systems represent the largest supercomputing installations in an academic setting.
Also new to the Top 10 is a Dell system at the National Centre for Supercomputing Applications (ranked eighth).
The fastest supercomputer in Europe is an IBM JS21 cluster at the Barcelona Supercomputing Center in Spain, which ranked ninth at 62.63 TFlop/s.
Rounding out the Top 10 is a new SGI system installed at the Leibniz Computer Center in Munich, with a performance of 56.52. TFlop/s.
The report saw the highest turnover rate in the project's 15 year history, with 284 systems falling off the end of the list.
The performance needed to make it onto the list increased to 4.005 TFlop/s compared to 2.737 TFlop/s six months ago.
Total combined performance for all 500 systems reached 4.92 PFlop/s (petaflops) compared to 3.54 PFlop/s six months ago and 2.79 PFlop/s just a year ago.
The system ranked number 500 on the current list would have held position number 216 only six months ago.
Geographically, the US remains the leading consumer of HPC systems with 281 of the 500 systems and the UK follows up in second place with 43 systems.
.jpg&h=140&w=231&c=1&s=0)
iTnews State of Security Breakfast
iTnews State of Data & AI Breakfast
Forrester's AI Forum Sydney
The 2026 iAwards
Security Exhibition & Conference
_(1).jpg&h=140&w=231&c=1&s=0)
