 |
The Cray XT3 is a distributed memory massively parallel MIMD supercomputer designed by Cray Inc. with Sandia National Laboratories under the codename Red Storm . Cray turned the design into a commercial product in 2004. The XT3 derives much of its architecture from the previous Cray T3E system, and also from the Intel ASCI Red supercomputer.
The XT3 consists of between 192 and 32,768 processing elements (PEs), where each PE comprises a 2.4 or 2.6 GHz AMD Opteron processor with up to two cores, a custom "SeaStar" communications chip, and between 1 and 8 GB of RAM. The PowerPC 440 based SeaStar device provides a 6.4 gigabyte per second connection to the processor across HyperTransport, as well as six 8-gigabyte per second links to neighboring PEs. The PEs are arranged in a 3-dimensional torus topology, with 96 PEs in each cabinet.
The XT3 runs an operating system called UNICOS/lc that partitions the machine into three sections, the largest comprising the Compute nodes, and two smaller sections for Service nodes and IO nodes. In UNICOS/lc 1.x, the Compute PEs run a Sandia developed microkernel called Catamount, which is descended from the SUNMOS OS of the Intel Paragon; in UNICOS/lc 2.0, Catamount was replaced by a specially tuned version of Linux called Compute Node Linux (CNL). Service and IO PEs run the full version of SuSE Linux and are used for interactive logins, systems management, application compiling and job launch. I/O PEs use physically distinct hardware, in that the node boards include PCI-X slots for connections to Ethernet and Fibre Channel networks.
Though the performance of each XT3 model will vary with the speed and number of processors installed, the November 2007 Top500 results for the Red Storm machine, the largest XT3 machine installed at Sandia, measured 102.7 teraflops on the Linpack benchmark, placing it at #6 on the list. After upgrades in 2008 to install some XT4 nodes with quad-core Opterons, Red Storm achieved 248 teraflops to place at #9 on the November 2008 Top500. The architecture was superseded in 2006 by the Cray XT4.
Cray computers | ||
|---|---|---|
| Cray Research |  | |
| Cray Computer Corp. | ||
| Cray Research Superservers | ||
| Cray Inc. | ||
UNICOS is a range of Unix-like operating system (OS) variants developed by Cray for its supercomputers. UNICOS is the successor of the Cray Operating System (COS). It provides network clustering and source code compatibility layers for some other Unixes. UNICOS was originally introduced in 1985 with the Cray-2 system and later ported to other Cray models. The original UNICOS was based on UNIX System V Release 2, and had many Berkeley Software Distribution (BSD) features added to it.
In computing, floating point operations per second is a measure of computer performance, useful in fields of scientific computations that require floating-point calculations. For such cases, it is a more accurate measure than measuring instructions per second.
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.
ASCI Red was the first computer built under the Accelerated Strategic Computing Initiative (ASCI), the supercomputing initiative of the United States government created to help the maintenance of the United States nuclear arsenal after the 1992 moratorium on nuclear testing.
SUNMOS is an operating system jointly developed by Sandia National Laboratories and the Computer Science Department at the University of New Mexico. The goal of the project, started in 1991, is to develop a highly portable, yet efficient, operating system for massively parallel-distributed memory systems.
The Cray T3E was Cray Research's second-generation massively parallel supercomputer architecture, launched in late November 1995. The first T3E was installed at the Pittsburgh Supercomputing Center in 1996. Like the previous Cray T3D, it was a fully distributed memory machine using a 3D torus topology interconnection network. The T3E initially used the DEC Alpha 21164 (EV5) microprocessor and was designed to scale from 8 to 2,176 Processing Elements (PEs). Each PE had between 64 MB and 2 GB of DRAM and a 6-way interconnect router with a payload bandwidth of 480 MB/s in each direction. Unlike many other MPP systems, including the T3D, the T3E was fully self-hosted and ran the UNICOS/mk distributed operating system with a GigaRing I/O subsystem integrated into the torus for network, disk and tape I/O.
Red Storm is a supercomputer architecture designed for the US Department of Energy’s National Nuclear Security Administration Advanced Simulation and Computing Program. Cray, Inc developed it based on the contracted architectural specifications provided by Sandia National Laboratories. The architecture was later commercially produced as the Cray XT3.
Roadrunner was a supercomputer built by IBM for the Los Alamos National Laboratory in New Mexico, USA. The US$100-million Roadrunner was designed for a peak performance of 1.7 petaflops. It achieved 1.026 petaflops on May 25, 2008, to become the world's first TOP500 LINPACK sustained 1.0 petaflops system.
HECToR was a British academic national supercomputer service funded by EPSRC, Natural Environment Research Council (NERC) and BBSRC for the UK academic community. The HECToR service was run by partners including EPCC, Science and Technology Facilities Council (STFC) and Numerical Algorithms Group (NAG).
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, a portable implementation of the high-performance LINPACK benchmark written in Fortran for distributed-memory computers.
The Bigben supercomputer was a Cray XT3 MPP system with 2068 nodes located at Pittsburgh Supercomputing Center. It was decommissioned on March 31, 2010. Bigben was a part of the TeraGrid.
The Cray XT5 is an updated version of the Cray XT4 supercomputer, launched on November 6, 2007. It includes a faster version of the XT4's SeaStar2 interconnect router called SeaStar2+, and can be configured either with XT4 compute blades, which have four dual-core AMD Opteron processor sockets, or XT5 blades, with eight sockets supporting dual or quad-core Opterons. The XT5 uses a 3-dimensional torus network topology.
The Cray CX1 is a deskside high-performance workstation designed by Cray Inc., based on the x86-64 processor architecture. It was launched on September 16, 2008, and was discontinued in early 2012. It comprises a single chassis blade server design that supports a maximum of eight modular single-width blades, giving up to 96 processor cores. Computational load can be run independently on each blade and/or combined using clustering techniques.
The National Center for Computational Sciences (NCCS) is a United States Department of Energy (DOE) Leadership Computing Facility that houses the Oak Ridge Leadership Computing Facility (OLCF), a DOE Office of Science User Facility charged with helping researchers solve challenging scientific problems of global interest with a combination of leading high-performance computing (HPC) resources and international expertise in scientific computing.
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.
A lightweight kernel (LWK) operating system is one used in a large computer with many processor cores, termed a parallel computer.
The Cray XT6 is an updated version of the Cray XT5 supercomputer, launched on 16 November 2009. The dual- or quad-core AMD Opteron 2000-series processors of the XT5 are replaced in the XT6 with eight- or 12-core Opteron 6100 processors, giving up to 2,304 cores per cabinet. The XT6 includes the same SeaStar2+ interconnect router as the XT5, which is used to provide a 3-dimensional torus network topology between nodes. Each XT6 node has two processor sockets, one SeaStar2+ router and either 32 or 64 GB of DDR3 SDRAM memory. Four nodes form one X6 compute blade.
The Cray XE6 made by Cray is an enhanced version of the Cray XT6 supercomputer, officially announced on 25 May 2010. The XE6 uses the same computer blade found in the XT6, with eight- or 12-core Opteron 6100 processors giving up to 3,072 cores per cabinet, but replaces the SeaStar2+ interconnect router used in the Cray XT5 and XT6 with the faster and more scalable Gemini router ASIC. This is used to provide a 3-dimensional torus network topology between nodes. Each XE6 node has two processor sockets and either 32 or 64 GB of DDR3 SDRAM memory. Two nodes share one Gemini router ASIC.
The Cray XK6 made by Cray is an enhanced version of the Cray XE6 supercomputer, announced in May 2011. The XK6 uses the same "blade" architecture of the XE6, with each XK6 blade comprising four compute "nodes". Each node consists of a 16-core AMD Opteron 6200 processor with 16 or 32 GB of DDR3 RAM and an Nvidia Tesla X2090 GPGPU with 6 GB of GDDR5 RAM, the two connected via PCI Express 2.0. Two Gemini router ASICs are shared between the nodes on a blade, providing a 3-dimensional torus network topology between nodes. This means that it has 576 GB of Graphics memory and over 1500 CPU cores, several orders of magnitude more powerful than the best publicly available computer on the market.
Catamount is an operating system for supercomputers.