LUMI

Last updated
LUMI
The LUMI supercomputer.jpg
ActiveJune 13, 2022
Sponsors European High-Performance Computing Joint Undertaking, LUMI Consortium
Location Kajaani, Finland
Architecture362,496 cores, AMD EPYC CPUs, 10,240 AMD Radeon Instinct MI250X GPUs (144,179,200 cores) [1] [2]
Power8.5 MW
Space150 m2
Memory1.75 petabytes
Storage117 petabytes
Speed550 petaFLOPS (peak)
Cost€144.5 million
Website www.lumi-supercomputer.eu

LUMI (Large Unified Modern Infrastructure) is a petascale supercomputer located at the CSC data center [3] in Kajaani, Finland. As of January 2023, the computer is the fastest supercomputer in Europe. [4]

Contents

The completed system consists of 362,496 cores, capable of executing more than 375 petaflops, with a theoretical peak performance of more than 550 petaflops, which places it among the top five most powerful computers in the world. [5] The November 2022 TOP500 ranks LUMI at number five, with a measured performance of 309.1 PFLOPS. [6]

Architecture

The system is being supplied by Hewlett Packard Enterprise (HPE), providing an HPE Cray EX supercomputer with next generation 64-core AMD EPYC CPUs and AMD Radeon Instinct GPUs. [7] [8] LUMI is a GPU based system, and the majority of its computing power comes from its GPU cores, an architecture which was chosen primarily for its cost/performance advantage. [9] The system is equipped with 1.75 petabytes of RAM, [1] [10] [11] and storage includes a 7-petabyte partition of flash storage, combined with 80-petabytes of traditional storage, both based on the Lustre parallel file system, as well as a 30-petabyte data management service based on Ceph. This gives the system a total of 117 petabytes of storage with an aggregated I/O bandwidth of 2 terabytes per second. [12]

Funding

LUMI is co-funded by the EuroHPC Joint Undertaking and the LUMI Consortium, which is composed of the following countries: Finland, Belgium, Czech Republic, Denmark, Estonia, Iceland, Norway, the Netherlands, Poland, Sweden, and Switzerland. The total budget is €144.5 million. [13] [14]

Energy

The computer uses 100% hydroelectric energy, and the heat it generates will be captured and used to heat buildings in the area, [15] [16] making LUMI one of the most environmentally efficient supercomputers in the world. [17] The former UPM paper mill where LUMI is located had only a single 2 minute power outage during its 38 years of operations thanks to the site's reliable connection to the national grid. [18]

Operation

Half of LUMI's capacity belongs to the EuroHPC Joint Undertaking, 20% of which is reserved for industry and SME use. [19] The other half is shared among the LUMI Consortium countries, according to each country’s financial contribution. [20]

By June 2021 pilot projects had been selected for the first run of the CPU partition, scheduled for September 2021, with full operations including the GPU partition planned for 2022. [21]

Naming

The word "lumi" means "snow" in Finnish. [22]

See also

Related Research Articles

Floating point operations per second is a measure of computer performance in computing, useful in fields of scientific computations that require floating-point calculations.

Cray Inc., a subsidiary of Hewlett Packard Enterprise, is an American supercomputer manufacturer headquartered in Seattle, Washington. It also manufactures systems for data storage and analytics. Several Cray supercomputer systems are listed in the TOP500, which ranks the most powerful supercomputers in the world.

<span class="mw-page-title-main">MareNostrum</span> Supercomputer in the Barcelona Supercomputing Center

MareNostrum is the main supercomputer in the Barcelona Supercomputing Center. It is the most powerful supercomputer in Spain, one of thirteen supercomputers in the Spanish Supercomputing Network and one of the seven supercomputers of the European infrastructure PRACE.

<span class="mw-page-title-main">TOP500</span> Database project devoted to the ranking of computers

The TOP500 project ranks and details the 500 most powerful non-distributed computer systems in the world. The project was started in 1993 and publishes an updated list of the supercomputers twice a year. The first of these updates always coincides with the International Supercomputing Conference in June, and the second is presented at the ACM/IEEE Supercomputing Conference in November. The project aims to provide a reliable basis for tracking and detecting trends in high-performance computing and bases rankings on HPL benchmarks, a portable implementation of the high-performance LINPACK benchmark written in Fortran for distributed-memory computers.

The Green500 is a biannual ranking of supercomputers, from the TOP500 list of supercomputers, in terms of energy efficiency. The list measures performance per watt using the TOP500 measure of high performance LINPACK benchmarks at double-precision floating-point format.

<span class="mw-page-title-main">Jaguar (supercomputer)</span> Cray supercomputer at Oak Ridge National Laboratory

Jaguar or OLCF-2 was a petascale supercomputer built by Cray at Oak Ridge National Laboratory (ORNL) in Oak Ridge, Tennessee. The massively parallel Jaguar had a peak performance of just over 1,750 teraFLOPS. It had 224,256 x86-based AMD Opteron processor cores, and operated with a version of Linux called the Cray Linux Environment. Jaguar was a Cray XT5 system, a development from the Cray XT4 supercomputer.

<span class="mw-page-title-main">Tianhe-1</span> Supercomputer

Tianhe-I, Tianhe-1, or TH-1 is a supercomputer capable of an Rmax of 2.5 peta FLOPS. Located at the National Supercomputing Center of Tianjin, China, it was the fastest computer in the world from October 2010 to June 2011 and was one of the few petascale supercomputers in the world.

Exascale computing refers to computing systems capable of calculating at least "1018 IEEE 754 Double Precision (64-bit) operations (multiplications and/or additions) per second (exaFLOPS)"; it is a measure of supercomputer performance.

<span class="mw-page-title-main">Supercomputing in Europe</span> Overview of supercomputing in Europe

Several centers for supercomputing exist across Europe, and distributed access to them is coordinated by European initiatives to facilitate high-performance computing. One such initiative, the HPC Europa project, fits within the Distributed European Infrastructure for Supercomputing Applications (DEISA), which was formed in 2002 as a consortium of eleven supercomputing centers from seven European countries. Operating within the CORDIS framework, HPC Europa aims to provide access to supercomputers across Europe.

The Cheyenne supercomputer at the NCAR-Wyoming Supercomputing Center (NWSC) in Cheyenne, Wyoming operated for seven years as one of the world's most powerful and energy-efficient computers from 2017 to 2024. Ranked in November 2016 as the 20th most powerful computer in the world and November 2023 as 159th by its TOP500 rank, the 5.34-petaflops system is capable of more than triple the amount of scientific computing performed by NCAR's previous supercomputer, Yellowstone. It also is three times more energy efficient than Yellowstone, with a peak computation rate of more than 3 billion calculations per second for every watt of energy consumed.

<span class="mw-page-title-main">Sierra (supercomputer)</span> Supercomputer developed by IBM

Sierra or ATS-2 is a supercomputer built for the Lawrence Livermore National Laboratory for use by the National Nuclear Security Administration as the second Advanced Technology System. It is primarily used for predictive applications in nuclear weapon stockpile stewardship, helping to assure the safety, reliability, and effectiveness of the United States' nuclear weapons.

<span class="mw-page-title-main">Frontier (supercomputer)</span> American supercomputer

Hewlett Packard Enterprise Frontier, or OLCF-5, is the world's first exascale supercomputer. It is hosted at the Oak Ridge Leadership Computing Facility (OLCF) in Tennessee, United States and became operational in 2022. As of December 2023, Frontier is the world's fastest supercomputer. It is based on the Cray EX and is the successor to Summit (OLCF-4). Frontier achieved an Rmax of 1.102 exaFLOPS, which is 1.102 quintillion floating-point operations per second, using AMD CPUs and GPUs.

The European High-Performance Computing Joint Undertaking is a public-private partnership in high-performance computing (HPC), enabling the pooling of European Union–level resources with the resources of participating EU Member States and participating associated states of the Horizon Europe and Digital Europe programmes, as well as private stakeholders. The Joint Undertaking has the twin stated aims of developing a pan-European supercomputing infrastructure, and supporting research and innovation activities. Located in Luxembourg City, Luxembourg, the Joint Undertaking started operating in November 2018 under the control of the European Commission and became autonomous in 2020.

Christofari — are Christofari (2019), Christofari Neo (2021) supercomputers of Sberbank based on Nvidia corporation hardware Sberbank of Russia and Nvidia. Their main purpose is neural network learning. They are also used for scientific research and commercial calculations.

<span class="mw-page-title-main">JUWELS</span> Supercomputer in Germany

JUWELS is a supercomputer developed by Atos and hosted by the Jülich Supercomputing Centre (JSC) of the Forschungszentrum Jülich.

Selene is a supercomputer developed by Nvidia, capable of achieving 63.460 petaflops, ranking as the fifth fastest supercomputer in the world, when it entered the list. Selene is based on the Nvidia DGX system consisting of AMD CPUs, Nvidia A100 GPUs, and Mellanox HDDR networking. Selene is based on the Nvidia DGX Superpod, which is a high performance turnkey supercomputer solution provided by Nvidia using DGX hardware. DGX Superpod is a tightly integrated system that combines high performance DGX compute nodes with fast storage and high bandwidth networking. It aims to provide a turnkey solution to high-demand machine learning workloads. Selene was built in three months and is the fastest industrial system in the US while being the second-most energy-efficient supercomputing system ever.

<span class="mw-page-title-main">Leonardo (supercomputer)</span> Supercomputer in Italy

Leonardo is a petascale supercomputer located at the CINECA datacenter in Bologna, Italy. The system consists of an Atos BullSequana XH2000 computer, with close to 14,000 Nvidia Ampere GPUs and 200 Gbit/s Nvidia Mellanox HDR InfiniBand connectivity. Inaugurated in November 2022, Leonardo is capable of 250 petaflops, making it one of the top five fastest supercomputers in the world. It debuted on the TOP500 in November 2022 ranking fourth in the world, and second in Europe.

HPC5 is a supercomputer built by Dell and installed by Eni, capable of 51.721 petaflops, and is ranked 9th in the Top500 as of November 2021. It is located in the Green Data Center in Ferrera Erbognone, in Northern Italy. In June 2020, HPC5 ranked 6th in the Green500. HPC5 is an upgrade to the HPC4 system, which was built by Hewlett Packard Enterprise and used by Eni. It is also called as HPC4+.

Voyager-EUS2 is a supercomputer built by Microsoft Azure, capable of 39.531 petaflops, and is ranked 14th in the TOP500 as of November 2022. Voyager-EUS2 runs from Microsoft Azure East US 2 region and it utilizes 253,440 cores on AMD EPYC CPUs along with an NVIDIA A100 GPU with 80GB memory and a Mellanox HDR Infiniband for data transfer running on Linux distribution.Si

Hewlett Packard Enterprise El Capitan, is an upcoming exascale supercomputer, hosted at the Lawrence Livermore National Laboratory in Livermore, United States and projected to become operational in 2024. It is based on the Cray EX Shasta architecture. When deployed, El Capitan is projected to displace Frontier as the world's fastest supercomputer.

References

  1. 1 2 "Archived copy". Archived from the original on 2023-03-31. Retrieved 2024-07-15.{{cite web}}: CS1 maint: archived copy as title (link)
  2. ↑ "Archived copy". Archived from the original on 2023-10-08. Retrieved 2024-07-15.{{cite web}}: CS1 maint: archived copy as title (link)
  3. ↑ "CSC: One of the world's mightiest supercomputers LUMI will lift European research and competitiveness to a new level". ScienceBusiness. Science Business Publishing International SRL. October 21, 2020. Archived from the original on July 15, 2024. Retrieved February 14, 2021.
  4. ↑ Larabel, Michael (13 June 2022). "LUMI Inaugurated As Europe's Most Powerful Supercomputer - Powered By AMD CPUs/GPUs". www.phoronix.com. Archived from the original on 14 June 2022. Retrieved 14 June 2022.
  5. ↑ Black, Doug (2020-05-01). "LUMI: the EuroHPC pre-exascale system of the North". insideHPC. Archived from the original on 2020-12-02. Retrieved 2021-02-15.
  6. ↑ "November 2022 | TOP500". www.top500.org. Archived from the original on 2022-11-16. Retrieved 2022-11-19.
  7. ↑ "Hewlett Packard Enterprise wins $160M+ contract to power one of the world's fastest supercomputers based in Finland to bolster Europe's research in science and unlock economic growth". Hewlett Packard Enterprise. Hewlett Packard Enterprise Development LP. October 21, 2020. Archived from the original on July 19, 2021. Retrieved February 13, 2021.
  8. ↑ Elster, Anne C. (February 26, 2021). "The European Factor: From ARM to Atos". Computing in Science & Engineering. 23 (1). IEEE: 102–105. Bibcode:2021CSE....23a.102E. doi:10.1109/MCSE.2020.3044070. S2CID   230608538. Archived from the original on 2024-07-15. Retrieved 2021-07-09.
  9. ↑ Maagaard Winther, Cecilie (December 6, 2019). "Huge HPC project offers great opportunities but is not without challenges". deic.dk. Danish e-infrastructure Cooperation. Retrieved February 15, 2021.
  10. ↑ "Archived copy". Archived from the original on 2023-03-31. Retrieved 2024-07-15.{{cite web}}: CS1 maint: archived copy as title (link)
  11. ↑ "Archived copy". Archived from the original on 2023-03-31. Retrieved 2024-07-15.{{cite web}}: CS1 maint: archived copy as title (link)
  12. ↑ "AMD's Next Gen EPYC & Radeon Instinct Powered LUMI Supercomputer Announced For 2021, 550 Petaflops Peak Horsepower". Wccftech.com. NewAge ADS, LLC. October 22, 2020. Archived from the original on July 15, 2024. Retrieved February 14, 2021.
  13. ↑ "LUMI: a new EuroHPC world-class supercomputer in Finland". EuroHPC. European Commission. October 21, 2020. Archived from the original on June 10, 2021. Retrieved February 14, 2021.
  14. ↑ "LUMI consortium". LUMI. Retrieved 2024-08-21.
  15. ↑ "LUMI Supercomputer Will Start Operations Next Year, Promotes Green Transition". HPCwire. Tabor Communications. November 17, 2020. Archived from the original on December 1, 2020. Retrieved February 13, 2021.
  16. ↑ "Be prepared for Euro-HPC resource LUMI | Karolinska Institutet". staff.ki.se. Archived from the original on 2021-07-09. Retrieved 2021-07-09.
  17. ↑ "LUMI supercomputer will now be built". Swedish Research Council. 27 October 2020. Archived from the original on 2021-07-09. Retrieved 2021-07-09.
  18. ↑ Manninen, Pekka (January 22, 2021). The pan-European supercomputer of the North (PDF) (Speech). 1st LUMI Roadshow. EuroCC National EuroHPC Competence Center Sweden. Archived (PDF) from the original on July 9, 2021. Retrieved July 9, 2021.
  19. ↑ Nordgren, Kaj (June 22, 2021). "Finland grants access to startups to supercomputer resources". Business Finland. Archived from the original on 2021-07-13. Retrieved 2021-07-13.
  20. ↑ "FAQ". LUMI. Archived from the original on 2021-07-13. Retrieved 2021-07-13.
  21. ↑ "LUMI pilot projects selected". LUMI. 2021-06-01. Archived from the original on 2021-07-12. Retrieved 2021-07-09.
  22. ↑ Alanne, Severi (1919). Suomalais-Englantilainen Sanakirja - Finnish-English Dictionary. Superior, Wisconsin: Tyomies Publishing Company. p. 343.
This page is based on this Wikipedia article
Text is available under the CC BY-SA 4.0 license; additional terms may apply.
Images, videos and audio are available under their respective licenses.