The Italian energy company Eni is upgrading its HPC4 system with new equipment from HPE, which is being installed in Eni’s Green Data Center in Ferrera Erbognone (a province in Pavia, Italy) and “as-a-service” via HPE’s GreenLake platform provided. The new HPC4, which includes more than 1,500 heterogeneous HPE ProLiant DL385 Gen10 Plus nodes, offers ~ 20 percent more computing power with a peak performance of around 22 petaflops.
The system’s 1,522 ProLiant nodes are equipped with two second-generation AMD Epyc “Rome” 24-core 7402 CPUs and a mix of GPU devices, all connected via Nvidia HDR InfiniBand. The majority of the system consists of 1,375 nodes that house two Nvidia “V100 O&G” GPUs; 125 nodes have two Nvidia A100s; and 22 nodes contain two AMD Instinct M100 GPUs. In addition, there are 20 pure CPU login nodes.
The new contract with HPE offers a complete update for HPC4, the HPE ProLiant DL380 Gen10 system installed in 2018. The advantages of the partners include an approximately 20 percent improvement in computing capacity for ongoing simulations and a doubling of the storage capacity to refine the accuracy of image-intensive modeling and simulations of complex energy research.
The new infrastructure uses the Cray ClusterStor E1000 storage system and the HPE Data Management Framework, “to support complex, image-intensive workloads in modeling and simulation”.
Eni told HPC wire that the combination of Eni Green Data Center and HPE GreenLake will increase sustainability by avoiding an oversupply of resources and the associated wasted energy consumption.
HPE said the new infrastructure will improve energy consumption and reduce e-waste by leveraging HPE Asset Upcycling Services. “That’s part of the Circular economy HPE Financial Services initiative that leverages asset longevity to reuse products by recycling equipment from the existing HPC4 system and replacing it with newer solutions, ”the company reported.
The original HPC4 installation (let’s call it HPC4-OG) was the largest energy supercomputer announced when it launched in 2018. It debuted at number 13 on the Top500 list in June 2018, delivering 12.2 Linpack petaflops out of 18.6 petaflops. HPC4-OG comprised 1,600 ProLiant DL380 nodes, each equipped with two Intel 24-core Skylake processors and two Nvidia Tesla P100 GPUs. The system 33rd place in the current Top500 list.
What about HPC5?
Eni still operates the HPC5 supercomputer, which was built by Dell and launched in February 2020. The HPC5 is housed in Eni’s Green Data Center and comprises 1,820 Dell EMC PowerEdge C4140 servers, each with two Intel Gold 6252 24-core processors and four Nvidia V100 GPUs. Server nodes are connected via Nvidia 200 Gbps HDR InfiniBand. The system was included in the Top500 list in June 2020 with 35.4 Linpack petaflops out of a theoretical 51.7 petaflops, making it the fastest industrial supercomputer in the world. Currently in ninth position on the list, HPC5 still holds this award.
In April last year, at the start of the pandemic when she devastated Italy, Eni announced that her HPC5 system was contributing to COVID-19 research as part of the European EXSCALATE4CoV project. Working with Cineca to bring its engineering skills and HPC5 computing power to the consortium, Eni enabled one of the most complex molecular supercomputing experiments ever conducted in November 2020.
Eni’s Green Data Center houses all of the company’s supercomputing systems and data. According to Eni, the Green Data Center was developed in order to achieve world-leading energy efficiency, partly due to the nearby photovoltaic system that supplies the system with part of its electricity.
HPE reports that its GreenLake business has a total contract value of more than $ 5.2 billion and approximately 1,200 enterprise customers in 50 countries. HPE recently won a GreenLake HPC-as-a-Service deal with the US National Security Agency (NSA) valued at $ 2 billion.