Misplaced Pages

This is an old revision of this page, as edited by 69.179.21.70 (talk) at 22:17, 21 April 2013. The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.

Titan is a supercomputer built by Cray at Oak Ridge National Laboratory for use in a variety of science projects. Titan is an upgrade of Jaguar, a previous supercomputer at Oak Ridge, to use graphics processing units (GPUs) in addition to conventional central processing units (CPUs), and is the first such hybrid to perform over 10 petaFLOPS. The upgrade began in October 2011, commenced stability testing in October 2012 and it became partially available to researchers in early 2013. The initial cost of the upgrade was US$60 million, funded primarily by the United States Department of Energy.

Titan has AMD Opteron CPUs in conjunction with Nvidia Tesla GPUs to improve energy efficiency while providing an order of magnitude increase in computational power over Jaguar. It uses 18,688 CPUs paired with an equal number of GPUs to perform at a theoretical peak of 27 petaFLOPS; however, in the LINPACK benchmark used to rank supercomputers' speed, it performed at 17.59 petaFLOPS. This was enough to take first place in the November 2012 list by the TOP500 organisation.

Titan is available for any purpose; however, selection for time on the computer depends on the importance of the project and its potential to fully utilise the hybrid architecture. Any selected code must also be executable on other supercomputers to avoid dependence solely on Titan. Six "vanguard" codes were selected to be the first to run on Titan dealing mostly with molecular scale physics or climate models, while 25 others are also queued for use of the machine. Due to the inclusion of GPUs, programmers have had to alter their existing code to properly address the new architecture. The modifications often require a greater degree of parallelism as the GPUs can handle many more threads simultaneously than CPUs and the changes often yield greater performance even on CPU-only machines.

History

Plans to create a supercomputer capable of 20 petaFLOPS at the Oak Ridge Leadership Computing Facility (OLCF) at Oak Ridge National Laboratory (ORNL) were in place as far back as 2005, when Jaguar was built. Titan will in-turn be replaced by a supercomputer designed to perform at approximately 200 petaFLOPS in 2016 as part of ORNL's plan to operate an exascale (1000 petaFLOPS to 1 exaFLOPS) machine by 2020. Initially a new 15,000 square metre (160,000 ft) building was planned to house the computer but it was eventually decided to use Jaguar's existing infrastructure. The precise nature of the hybrid architecture was not finalised until 2010 although a deal with Nvidia to supply the GPUs was signed in 2009. Titan was first announced at the private ACM/IEEE Supercomputing Conference (SC10) on November 16, 2010 and was publicly announced on October 11, 2011 as the first phase of the upgrade began.

In order to remain up to date with power efficiency and processing power, Jaguar had received various upgrades since its creation in 2005 when it used the Cray XT3 platform and performed at 25 teraFLOPS. By 2008, Jaguar had been expanded with more cabinets and upgraded to the XT4 platform so that it performed at 263 teraFLOPS. In 2009, it was upgraded to the XT5 platform and performed at 1.4 petaFLOPS. Further upgrades brought Jaguar to 1.76 petaFLOPS before the Titan upgrades began.

Titan was funded primarily by the US Department of Energy through Oak Ridge National Laboratory. ORNL funding was sufficient to purchase the CPUs but not all of the GPUs so the National Oceanic and Atmospheric Administration agreed to fill the remaining nodes in return for computing time. ORNL scientific computing chief Jeff Nichols noted that Titan cost approximately $60 million upfront, of which the NOAA contribution was less than $10 million, but would not release precise figures due to non-disclosure agreements. Throughout the full term of the contract with Cray, Titan will cost $97 million not including potential upgrades to the machine.

Jaguar's internals were upgraded over the course of a year beginning October 9, 2011. Between October and December, 96 of Jaguar's 200 cabinets, each containing 24 XT5 blades (two 6-core CPUs per node, four nodes per blade), were upgraded to XK7 blades (one 16-core CPU per node, still four nodes per blade) while the remainder of the machine was still available for processing. In December, computation was moved to the 96 XK7 cabinets and the remaining 104 cabinets were upgraded to XK7 blades. The system's interconnect (the network that allows the CPUs to communicate with each other) was updated and the ORNL's ESnet connection was upgraded to 100 Gbit/s to permit faster data transfer to other national laboratories, universities and research institutions. The system memory was doubled to 600 TB as the nodes were upgraded to XK7. 960 of the XK7 nodes (10 cabinets) were also fitted with a Fermi based GPU as Kepler GPUs were not then available; these 960 nodes were referred to as TitanDev and used to test code for Titan's full upgrade. This first phase of the upgrade increased the peak performance of Jaguar to 3.3 petaFLOPS. Beginning on September 13, 2012, Nvidia K20X GPUs were fitted to all of Jaguar's XK7 compute blades, including the 960 TitanDev nodes. In October, the task was completed and the computer was renamed Titan, having remained "Jaguar" until that point.

Titan underwent acceptance testing in early 2013 but only completed 92% of the tests, short of the required 95%. The problem was discovered to be an excess of gold in the female edge connectors of the motherboards' PCIe slots causing cracks in the motherboards' solder. The cost of repair was borne by Cray and between 12 and 16 cabinets were repaired each week. Throughout the repairs researchers were given access to the available CPUs, and on March 11, researchers were granted access to 8,972 GPUs. ORNL announced on April 8 that the repairs were complete, and expects acceptance testing to be completed in May 2013.

In March 2013, Nvidia launched the GTX Titan, a consumer graphics card that uses the same GPU die as the K20X GPUs in Titan. The relation to the supercomputer is prominent in marketing of the GTX Titan.

Hardware

Titan uses the same 200 cabinets covering 404 square metres (4,352 ft) that Jaguar did, with replaced internals and upgraded networking facilities. Reusing the power and cooling systems already in place for Jaguar saved the lab approximately US$20 million and power is provided to each cabinet at 480 V to use thinner cables than the US standard 208 V, saving US$1 million in copper. At its peak, Titan draws 8.2 MW, 1.2 MW more than Jaguar did, but it is almost ten times as fast in terms of floating point calculations. In the event of a power failure, carbon fibre flywheels power generators can keep the networking and storage infrastructure running for up to 16 seconds. If power is not restored within 2 seconds, diesel engines are started, taking approximately 7 seconds to come up to full power, and assuming the role of powering the generators indefinitely. The generators are designed only to keep the networking and storage components powered so that a reboot is much quicker; the generators are not capable of powering the processing infrastructure to continue simulations. Titan's components are air-cooled by heatsinks, but the air is chilled before being pumped through the cabinets. The noise of the fans cooling the components is so loud that hearing protection is required for people spending more than 15 minutes in the same room as the machine. The cooling system has a cooling capacity of 6000 tonnes (6600 tons) and works by chilling water to 5.5 °C (42 °F), which in turn cools recirculated air.

Titan has 18,688 nodes (4 nodes per blade, 24 blades per cabinet), each containing a 16-core AMD Opteron 6274 CPU with 32 GB of DDR3 ECC memory and an Nvidia Tesla K20X GPU with 6 GB GDDR5 ECC memory. There are a total of 299,008 processor cores and just over 710 TB of RAM. Initially Titan used the 10 PB of storage with a transfer speed of 240 GB/s that Jaguar had, but in April 2013, the storage was upgraded to 40 PB with a transfer of 1.4 TB/s. Titan runs the Cray Linux Environment, a full version of Linux on the login nodes that users directly access, but a scaled down, more efficient version on the compute nodes. GPUs were selected for their vastly higher parallel processing efficiency over CPUs. Although the GPUs have a slower clock speed than the CPUs, each GPU contains 2,688 CUDA cores at 732 MHz, resulting in a faster overall system. Consequently, the CPUs' cores are used to allocate tasks to the GPUs rather than directly processing the data as in conventional supercomputers.

Titan's hardware has a theoretical peak performance of 27 petaFLOPS with perfectly optimised software. On November 12, 2012 the TOP500 organisation that ranks the worlds' supercomputers by their LINPACK performance, announced that Titan was ranked first at 17.59 petaFLOPS, displacing IBM Sequoia. Titan was also ranked third on the Green500, the same 500 supercomputers re-ordered in terms of energy efficiency. Researchers also have access to EVEREST (Exploratory Visualisation Environment for Research and Technology) to better understand the data that Titan outputs. EVEREST is a visualisation room with a 10 by 3 metre (33 by 10 ft) screen and a smaller, secondary screen. The screens are 37 and 33 megapixels respectively with stereoscopic 3D capability.

Research projects

Although Titan is available for use by any project, the requests for use exceeded the computing time available, so selection criteria were drawn up. In 2009, the Oak Ridge Leadership Computing Facility considered fifty applications for first use of the supercomputer, but narrowed it down to six successful "vanguard" codes chosen not only for the importance of the research, but for their ability to fully utilise the computing power of the hybrid system. Thirty-one codes will be run on Titan throughout 2013, typically four or five at any one time. The code of many projects had to be modified to suit the GPU processing of Titan, but each code was required to still be capable of running on CPU-based systems so that the projects were not solely dependent on Titan. OLCF formed the Center for Accelerated Application Readiness (CAAR) to aid researchers in modifying their code for Titan and holds developer workshops at Nvidia headquarters to educate users about the architecture, compilers and applications on Titan. CAAR has been working on compilers with Nvidia and code vendors to integrate directives for GPUs into their programming languages. Researchers can thus express parallelism in their code with their existing programming language, typically Fortran, C or C++, and the compiler can express it to the GPUs. Dr. Bronson Messer, a computational astrophysicist, said of the task: "...an application using Titan to the utmost must also find a way to keep the GPU busy, remembering all the while that the GPU is fast, but less flexible than the CPU." Moab Cluster Suite is used to prioritise jobs to nodes to keep utilisation high; it improved efficiency from 70% to approximately 95% in the tested software. Some projects found that the changes increased efficiency of their code on non-GPU machines; the performance of Denovo doubled on CPU-based machines.

The six vanguard projects to use Titan are:

• S3D, a project that models the molecular physics of combustion, aims to improve the efficiency of diesel and biofuel engines. In 2009, they produced the first fully resolved simulation of autoigniting hydrocarbon flames relevant to the efficiency of direct injection diesel engines using Jaguar.
• WL-LSMS simulates the interactions between electrons and atoms in magnetic materials at temperatures other than absolute zero. An earlier version of the code was the first to perform at greater than one petaFLOPS on Jaguar.
• Denovo simulates nuclear reactions with the aim of improving the efficiency and reducing the waste of nuclear reactors. The performance of Denovo on conventional CPU-based machines doubled after the tweaks for Titan and performs 3.5 faster on Titan than it did on Jaguar.
• Large-scale Atomic/Molecular Massively Parallel Simulator (LAMMPS) is a molecular dynamics code that simulates particles across a range of scales, from quantum to relativistic, to improve materials science with potential applications in semi-conductor, biomolecule and polymer development.
• CAM-SE is a combination of two codes: Community Atmosphere Model, a global atmosphere model, and High Order Method Modeling Environment, a code that solves fluid and thermodynamic equations. CAM-SE will allow greater accuracy in climate simulations.
• Non-Equilibrium Radiation Diffusion (NRDF) plots non-charged particles through supernovae with potential applications in laser fusion, fluid dynamics, medical imaging, nuclear reactors, energy storage and combustion.

The amount of code alteration required to run on the GPUs varies by project. According to Dr. Messer of the NRDF project, only a small percentage of his code is for the GPUs because the calculations are relatively simple but processed repeatedly and in parallel. NRDF is written in CUDA Fortran, a version of normal Fortran with CUDA extensions for the GPUs. Dr. Messer's research requires hundreds of partial differential equations to track the energy, angle, angle of scatter and type of each neutrino modeled in a star going supernova, resulting in millions of individual equations. The code was named Chimera after the mythological creature because it has three "heads": the first simulates the hydrodynamics of stellar material, the second simulates radiation transport and the third simulates nuclear burning. The third "head" is the first to run on the GPUs as the nuclear burning can most easily be simulated by GPU architecture although the other aspects of the code will be modified in time. On Jaguar, the project models 14 or 15 nuclear species but Dr. Messer anticipates up to 200 species could be simulated, allowing far greater precision when comparing the simulation to empirical observation.

VERA is a light water reactor simulation written at the Consortium for Advanced Simulation of Light Water Reactors (CASL) on Jaguar. VERA allows engineers to monitor the performance and status of any part of a reactor core throughout the lifetime of the reactor to identify points of interest. Although not one of the first six projects, VERA will be run on Titan having been optimised with assistance from CAAR and tested on TitanDev. Computer scientist Tom Evans found that the adaption to Titan's hybrid architecture was of greater difficulty than between previous CPU based supercomputers. Despite this, he aims to simulate an entire reactor fuel cycle, an eighteen to thirty-six month process, in one week on Titan.

See also

• List of Advanced Scientific Computing Research Leadership Computing Challenge allocations
• Oak Ridge Leadership Computing Facility

References

1. ^ "Oak Ridge Claims No. 1 Position on Latest TOP500 List with Titan". TOP500. November 12, 2012. Archived from the original on January 21, 2013. Retrieved November 15, 2012.
2. ^ "Discussing the ORNL Titan Supercomputer with ORNL's Jack Wells". The Exascale Report. 2012. Archived from the original on March 26, 2013. Retrieved December 19, 2012. {{cite web}}: Italic or bold markup not allowed in: |publisher= (help); Unknown parameter |month= ignored (help)
3. ^ Bland, Buddy (November 16, 2010). "Where do we go from here?" (PDF). Oak Ridge National Laboratory. Archived from the original (PDF) on March 3, 2012. Retrieved December 18, 2012.
4. Goldman, David (October 29, 2012). "Top U.S. supercomputer guns for fastest in world". CNN. Archived from the original on March 2, 2013. Retrieved March 31, 2013. {{cite web}}: Italic or bold markup not allowed in: |publisher= (help)
5. Munger, Frank (March 7, 2011). "Oak Ridge lab to add titanic supercomputer". Knox News. Archived from the original on July 4, 2012. Retrieved December 19, 2012. {{cite web}}: Italic or bold markup not allowed in: |publisher= (help)
6. Morgan, Timothy Prickett (October 1, 2009). "Oak Ridge goes gaga for Nvidia GPUs". The Register. Archived from the original on November 9, 2012. Retrieved December 19, 2012. {{cite web}}: Italic or bold markup not allowed in: |publisher= (help)
7. Levy, Dawn (October 11, 2011). "ORNL awards contract to Cray for Titan supercomputer". Oak Ridge National Laboratory. Archived from the original on February 26, 2013. Retrieved December 19, 2012.
8. ^ "Jaguar: Oak ridge National Laboratory". TOP500. Archived from the original on March 17, 2013. Retrieved December 18, 2012.
9. "TOP500 List November 2011". TOP500. Archived from the original on January 21, 2013. Retrieved December 18, 2012.
10. ^ Munger, Frank (November 26, 2012). "The ORNL and NOAA relationship". Knox News. Archived from the original on March 26, 2013. Retrieved December 20, 2012. {{cite web}}: Italic or bold markup not allowed in: |publisher= (help)
11. Munger, Frank (November 18, 2012). "The cost of Titan". Knox News. Archived from the original on March 26, 2013. Retrieved December 20, 2012. {{cite web}}: Italic or bold markup not allowed in: |publisher= (help)
12. ^ Feldman, Michael (October 11, 2011). "GPUs Will Morph ORNL's Jaguar Into 20-Petaflop Titan". HPC Wire. Archived from the original on July 27, 2012. Retrieved October 29, 2012. {{cite web}}: Italic or bold markup not allowed in: |publisher= (help)
13. ^ "Titan Project Timeline". Oak Ridge Leadership Computing Facility. Archived from the original on June 18, 2012. Retrieved December 18, 2012.
14. ^ Brouner, Jennifer; McCorkle, Morgan; Pearce, Jim; Williams, Leo (2012). "ORNL Review Vol. 45" (PDF). Oak Ridge National Laboratory. Archived from the original (PDF) on March 4, 2013. Retrieved November 2, 2012.
15. "Superfast Titan, Superfast Network". Oak Ridge Leadership Computing Facility. December 17, 2012. Archived from the original on March 26, 2013. Retrieved December 18, 2012.
16. Poeter, Damon (October 11, 2011). "Cray's Titan Supercomputer for ORNL Could Be World's Fastest". PC Magazine. Archived from the original on June 5, 2012. Retrieved October 29, 2012. {{cite web}}: Italic or bold markup not allowed in: |publisher= (help)
17. Jones, Gregory Scott (September 17, 2012). "Final Upgrade Underway". Oak Ridge Leadership Computing Facility. Archived from the original on March 26, 2013. Retrieved November 16, 2012.
18. Munger, Frank (February 20, 2013). "No. 1 Titan not yet living up to potential". Knox News. Archived from the original on March 26, 2013. Retrieved March 26, 2013. {{cite web}}: Italic or bold markup not allowed in: |publisher= (help)
19. ^ Huotari, John (March 13, 2013). "Cray re-soldering Titan's connectors, supercomputer testing could be done in April". Oak Ridge Today. Archived from the original on March 26, 2013. Retrieved March 26, 2013. {{cite web}}: Italic or bold markup not allowed in: |publisher= (help)
20. Jones, Scott (March 26, 2013). "Titan Users Now Have Access to GPUs". Oak Ridge Leadership Computing Facility. Archived from the original on March 13, 2013. Retrieved March 26, 2013.
21. Huotari, John (April 8, 2013). "Titan repairs complete, ORNL preparing for second round of supercomputer testing". Oak Ridge Today. Archived from the original on April 8, 2013. Retrieved April 8, 2013. {{cite web}}: Italic or bold markup not allowed in: |publisher= (help)
22. Smith, Ryan (February 21, 2013). "Nvidia's GeForce GTX Titan Review, Part 2: Titan's Performance Unveiled". Anandtech. Archived from the original on February 23, 2013. Retrieved March 26, 2013. {{cite web}}: Italic or bold markup not allowed in: |publisher= (help)
23. "GeForce GTX Titan". Nvidia. Archived from the original on March 26, 2013. Retrieved March 26, 2013.
24. ^ Tibken, Shara (October 29, 2012). "Titan supercomputer debuts for open scientific research". CNET. Archived from the original on December 15, 2012. Retrieved October 29, 2012. {{cite web}}: Italic or bold markup not allowed in: |publisher= (help)
25. ^ "Introducing Titan". Oak Ridge Leadership Computing Facility. Archived from the original on February 22, 2013. Retrieved October 29, 2012.
26. Munger, Frank (October 29, 2012). "Titan's ready to roll; ORNL supercomputer may become world's No. 1". Knox News. Archived from the original on March 26, 2013. Retrieved October 29, 2012. {{cite web}}: Italic or bold markup not allowed in: |publisher= (help)
27. ^ Lal Shimpi, Anand (October 31, 2012). "Inside the Titan Supercomputer". Anandtech. p. 1. Archived from the original on January 25, 2013. Retrieved November 2, 2012. {{cite web}}: Italic or bold markup not allowed in: |publisher= (help)
28. "Heterogeneous Systems Re-Claim Green500 List Dominance". Green500. November 14, 2012. Archived from the original on February 5, 2013. Retrieved November 15, 2012.
29. ^ Bland, Buddy; Lal Shimpi, Anand (October 30, 2012). "Oak Ridge National Laboratory Tour – Backup Power" (Youtube). Anandtech. Retrieved November 2, 2012. {{cite web}}: Italic or bold markup not allowed in: |publisher= (help)
30. ^ Bland, Buddy; Lal Shimpi, Anand (October 30, 2012). "Oak Ridge National Laboratory Tour – Cooling Requirements" (Youtube). Anandtech. Retrieved November 2, 2012. {{cite web}}: Italic or bold markup not allowed in: |publisher= (help)
31. ^ Pavlus, John (October 29, 2012). "Building Titan: The 'world's fastest' supercomputer". BBC. Archived from the original on January 20, 2013. Retrieved January 8, 2013. {{cite web}}: Italic or bold markup not allowed in: |publisher= (help)
32. Morgan, Timothy Prickett (October 11, 2011). "Oak Ridge changes Jaguar's spots from CPUs to GPUs". The Register. Archived from the original on October 15, 2012. Retrieved December 21, 2012. {{cite web}}: Italic or bold markup not allowed in: |publisher= (help)
33. ^ "ORNL Debuts Titan Supercomputer" (PDF). Oak Ridge Leadership Computing Facility. Archived from the original (PDF) on February 26, 2013. Retrieved October 29, 2012.
34. Lal Shimpi, Anand (October 31, 2012). "Titan's storage array". Anandtech. Archived from the original on March 26, 2013. Retrieved December 18, 2012. {{cite web}}: Italic or bold markup not allowed in: |publisher= (help)
35. Santos, Alexis (April 16, 2013). "Titan supercomputer to be loaded with 'world's fastest' storage system". Engadget. Archived from the original on April 16, 2013. Retrieved April 16, 2013.
36. "Titan System Overview". Oak Ridge Leadership Computing Facility. Archived from the original on March 26, 2013. Retrieved December 21, 2012.
37. Smith, Ryan (November 12, 2012). "NVIDIA Launches Tesla K20 & K20X: GK110 Arrives At Last". Anandtech. Archived from the original on January 24, 2013. Retrieved December 21, 2012. {{cite web}}: Italic or bold markup not allowed in: |publisher= (help)
38. Feldman, Michael (October 29, 2012). "Titan Sets High Water Mark for GPU Supercomputing". HPC Wire. Archived from the original on March 26, 2013. Retrieved October 30, 2012. {{cite web}}: Italic or bold markup not allowed in: |publisher= (help)
39. Jones, Gregory Scott (November 12, 2012). "ORNL Supercomputer Named World's Most Powerful". Oak Ridge National Laboratory. Archived from the original on February 22, 2013. Retrieved December 14, 2012.
40. "US Titan supercomputer clocked as world's fastest". BBC. November 12, 2012. Archived from the original on February 3, 2013. Retrieved November 12, 2012. {{cite web}}: Italic or bold markup not allowed in: |publisher= (help)
41. Williams, Leo (November 14, 2012). "Titan is Also a Green Powerhouse". Oak Ridge Leadership Computing Facility. Archived from the original on February 16, 2013. Retrieved November 15, 2012.
42. Munger, Frank (January 1, 2013). "ORNL visualization lab gets $2.5M makeover, adds 3D". Knox News. Archived from the original on March 26, 2013. Retrieved January ,2013. {{cite web}}: Check date values in: |accessdate= (help); Italic or bold markup not allowed in: |publisher= (help)
43. ^ "TITAN: Built for Science" (PDF). Oak Ridge Leadership Computing Facility. Archived from the original (PDF) on February 26, 2013. Retrieved October 29, 2012.
44. "2013 INCITE Awards". US Department of Energy. Archived from the original (PDF) on March 26, 2013. Retrieved January 17, 2013.
45. ^ Williams, Leo. "Preparing users for Titan". Oak Ridge National Laboratory. Archived from the original on March 1, 2013. Retrieved November 19, 2012.
46. Rumsey, Jeremy (December 17, 2012). "Titan Trainers Take Road Trip". Oak Ridge Leadership Computing Facility. Archived from the original on March 26, 2013. Retrieved December 18, 2012.
47. "Supercomputing Leaders Choose Adaptive Computing to Complement Latest HPC Systems". Business Wire. January 30, 2013. Archived from the original on March 26, 2013. Retrieved January 31, 2013. {{cite web}}: Italic or bold markup not allowed in: |publisher= (help)
48. DuBois, Shelley (January 30, 2013). "The next revolution in cloud computing". Fortune Magazine. Archived from the original on March 26, 2013. Retrieved January 31, 2013. {{cite web}}: Italic or bold markup not allowed in: |publisher= (help)
49. "Nuclear Energy – Supercomputer speeds path forward". Consortium for Advanced Simulation of LWRs. Archived from the original on February 14, 2013. Retrieved December 14, 2012.
50. Zybin, Sergey. "LAMMPS Molecular Dynamics Simulator". Sandia National Laboratories. Archived from the original on February 16, 2013. Retrieved October 29, 2012.
51. ^ Lal Shimpi, Anand (October 31, 2012). "Inside the Titan Supercomputer". Anandtech. p. 3. Archived from the original on March 26, 2013. Retrieved November 15, 2012. {{cite web}}: Italic or bold markup not allowed in: |publisher= (help)
52. ^ Messer, Bronson (October 30, 2012). "Using Titan to Model Supernovae" (Youtube). Anandtech. Retrieved November 15, 2012. {{cite web}}: Italic or bold markup not allowed in: |publisher= (help)
53. ^ Pearce, Jim. "VERA analyzes nuclear reactor designs in unprecedented detail". Oak Ridge National Laboratory. Archived from the original on February 15, 2013. Retrieved December 18, 2012.

External links

• Official website
• Media related to Titan (supercomputer) at Wikimedia Commons

• Cray
• Supercomputers
• Oak Ridge National Laboratory
• One-of-a-kind computers
• Nvidia