_logo_2017.svg){kind=logo} | |
| Sponsors | United States Department of Energy |
|---|---|
| Operators | IBM |
| Architecture | 9,216POWER9 22-core CPUs 27,648Nvidia Tesla V100 GPUs[1] |
| Power | 13MW[2] |
| Operating system | Red Hat Enterprise Linux (RHEL)[3][4] |
| Storage | 250PB |
| Speed | 200petaFLOPS (peak) |
| Ranking | TOP500: 7 (1H2024) |
| Purpose | Scientific research |
| Website | www |
.jpg)
Summit orOLCF-4 was asupercomputer developed byIBM for use atOak Ridge Leadership Computing Facility (OLCF), a facility at theOak Ridge National Laboratory, United States of America. It held the number 1 position on theTOP500 list from June 2018 to June 2020.[5][6] As of June 2024, itsLINPACK benchmark was clocked at 148.6 petaFLOPS.[7] Summit was decommissioned on November 15, 2024.[8]
As of November 2019, the supercomputer had ranked as the 5th most energy efficient in the world with a measured power efficiency of 14.668 gigaFLOPS/watt.[9] Summit was the first supercomputer to reach exaflop (a quintillion operations per second) speed, on a non-standard metric, achieving 1.88 exaflops during agenomic analysis and is expected to reach 3.3 exaflops usingmixed-precision calculations.[10]
TheUnited States Department of Energy awarded a $325 million contract in November 2014 to IBM,Nvidia andMellanox. The effort resulted in construction of Summit andSierra. Summit is tasked with civilian scientific research and is located at the Oak Ridge National Laboratory in Tennessee. Sierra is designed for nuclear weapons simulations and is located at theLawrence Livermore National Laboratory in California.[11]
Summit was estimated to cover 5,600 square feet (520 m2)[12] and require 219 kilometres (136 mi) of cabling,[13] and was designed to be used for research in diverse fields such ascosmology,medicine, andclimatology.[14]
In 2015, the project called Collaboration of Oak Ridge, Argonne and Lawrence Livermore (CORAL) included a third supercomputer namedAurora and was planned for installation atArgonne National Laboratory.[15] By 2018, Aurora was re-engineered with completion anticipated in 2021 as anexascale computing project along withFrontier andEl Capitan to be completed shortly thereafter.[16] Aurora was completed in late 2022.[17]
The Summit supercomputer was built for research in energy, artificial intelligence, human health, and other areas.[18] It has been used in earthquake simulation, extreme weather simulation, materials science, genomics, and predicting the lifetime of neutrinos.[19]
 | This sectionmay be too technical for most readers to understand. Pleasehelp improve it tomake it understandable to non-experts, without removing the technical details.(May 2020) (Learn how and when to remove this message) |
Each of its 4,608 nodes consist of 2 IBMPOWER9CPUs, 6Nvidia TeslaGPUs,[20] with over 600 GB ofcoherent memory (96 GBHBM2 plus 512 GBDDR4) which is addressable by all CPUs and GPUs, plus 800 GB ofnon-volatile RAM that can be used as aburst buffer or as extended memory.[21] ThePOWER9 CPUs andNvidia Volta GPUs are connected using Nvidia's high speedNVLink. This allows for aheterogeneous computing model.[22]
To provide a high rate of data throughput, the nodes are connected in a non-blockingfat-tree topology using a dual-rail Mellanox EDRInfiniBand interconnect for both storage and inter-process communications traffic, which delivers both 200 Gbit/s bandwidth between nodes and in-network computing acceleration for communications frameworks such asMPI andSHMEM/PGAS.
The storage for Summit[23] has a fast in-system layer and a center-wide parallel filesystem layer. The in-system layer is optimized for fast storage with SSDs on each node, while the center-wide parallel file system provides easy to access data stored on hard drives. The two layers work together seamlessly so users do not have to differentiate their storage needs. The center-wide parallel file system isGPFS (IBM Storage Scale). It provides 250PB of storage. The cluster delivers 2.5 TB/s of single stream read peak throughput and 1 TB/s of 1M file throughput. It was one of the first supercomputers that also required extremely fast metadata performance to support AI/ML workloads exemplified by the 2.6M 32k file creates per second it delivers.
| Records | ||
|---|---|---|
| Preceded by Sunway TaihuLight 93.01 petaFLOPS | World's most powerful supercomputer June 2018 - June 2020 148.6 petaFLOPS | Succeeded by RIKEN Fugaku 0.54 exaFLOPS |