The Italian company Eni is revolutionizing the world of scientific computing with the supercomputer HPC6, an unparalleled technological feat in Europe. This digital giant, equipped with nearly 14,000 GPUs, allows for the analysis of massive data sets and shapes the future of energy. This model offers phenomenal performance, propelling Eni to fifth place globally among the most powerful supercomputers.
The quest for an energy transition towards more sustainable solutions no longer seems out of reach. Thanks to this colossal investment of $104.1 million, Eni arms itself with essential tools to locate new oil and gas reserves, while enhancing its efforts towards green energy.
The Eni HPC6: a technological leap
The Italian energy company Eni recently commissioned HPC6, its revolutionary supercomputer. This data processing system, equipped with nearly 14,000 GPUs, is dedicated to analyzing large data sets. Its main objective: to discover new gas and oil reserves. Ranked among the most powerful in the world, it holds the fifth place with a capacity of 477 million trillion operations per second.
Performance and technical characteristics
HPC6 is installed at the Ferrera Erbognone data center, near Milan. This supercomputer relies on a direct liquid cooling system, which improves its energy efficiency. Its architecture includes 3,472 computing units, each integrating an AMD EPYC processor with 64 cores, paired with four AMD Instinct MI250X GPUs.
In total, HPC6 integrates 13,888 GPUs for peak performance. This computing power, nine times higher than its predecessor, the HPC5, allows it to reach 606 petaflops at peak and 477 petaflops continuously.
Industrial and scientific applications
One of the main uses of HPC6 involves discovering new materials. This supercomputer will also analyze plasma clouds in nuclear fusion reactors, while improving technologies related to carbon emission capture. Eni aspires to develop more efficient solar panels thanks to this new infrastructure.
This innovative machine also promises to accelerate research in the field of green energy and decarbonization. The company aims to adapt its research protocols to meet current environmental challenges.
Commitment to the future
Eni, traditionally focused on hydrocarbons, is now making a shift towards renewable energies. Research dedicated to these clean sources occupies 70% of its teams’ time. This strategic change involves continuous integration of new technologies to optimize the energy transition.
This proactive approach positions Eni as a key player in the fight against climate change and sustainable development while innovating in the fossil fuels sector. This cutting-edge supercomputer represents a catalyst on this path.
Frequently Asked Questions
What is the name of the supercomputer commissioned by Eni?
The supercomputer commissioned by Eni is called HPC6.
How many GPUs does the HPC6 supercomputer have?
The HPC6 is equipped with 13,888 GPUs.
What is the main use of the HPC6 supercomputer?
The HPC6 is primarily used to analyze large amounts of data in order to discover new gas and oil reserves.
Where is Eni’s HPC6 supercomputer installed?
The HPC6 supercomputer is installed in the Ferrera Erbognone data center, located 50 kilometers southwest of Milan.
What is the computing capacity of the HPC6?
The HPC6 achieves a capacity of 477 petaflops continuously and shows peak performance of 606 petaflops.
What is the commissioning cost of the HPC6 supercomputer?
The commissioning cost of the HPC6 supercomputer is estimated at $104.1 million (€99.6 million).
What are the HPC6’s features related to artificial intelligence?
The HPC6 was designed to support various workloads related to artificial intelligence, allowing for complex analyses and advanced simulations.
How does Eni plan to use the HPC6 for green energy research?
Eni plans to use the HPC6 to accelerate research on green energy and improve decarbonization technologies.
What is the performance difference between the HPC6 and its predecessor, the HPC5?
The HPC6 is nine times faster than the HPC5, which showed a speed of 51.7 petaflops.
