Silicon Graphics Inc (SGI^TM) is the world's leader in high-performance computing technology. The company's systems, ranging from workstations and servers to the most powerful supercomputers in the world, deliver advanced computing and 3D visualization capabilities to scientific, engineering, and creative professionals and large enterprises. 

Servers:

SGI^TM servers are the market leaders in technical computing applications, empowering insight in key industries such as manufacturing, government, entertainment, communications, energy, science, and education. The company's highly scalable servers also have a growing presence in the commercial market, with an emphasis on strategic business analysis, Internet applications, and digital media serving. SGI^TM servers provide a time reliable centralized computing asset. Whether you need solutions for manufacturing, science and medicine, telecommunications or for enterprise, the Challenge^TM, Origin^TM, and SGI 1400 server families scale to accommodate your strategic computing needs. The scalability and absolute power of UNIX-based Challenge and Origin servers handle the most demanding mission-critical applications worldwide. The SGI 1400 servers, employing Intel® Pentium® III Xeon^TM processors, deliver SGI's proven high-performance and scalability as well as affordability and the reliability of open server solutions.

The Challenge family offers a choice of servers spanning every level of price/performance for any workgroup, department, or enterprise environment. The entry level member of the Challenge family is the Challenge S, a high performance desktop server. The Challenge DM is the entry into our deskside symmetrical multiprocessing servers. The Challenge L is a more powerful deskside SMP server for compute intensive applications. At the high-end, the Challenge XL offers the most expandable and powerful Unix-based SMP server in the industry.

Challenge S

The Challenge S is the entry-level member of the CHALLENGE family of servers. It provides tremendous performance in a compact chassis with leading edge I/O performance, and is an ideal choice for small to mid-size workgroups, and World Wide Web Serving.

The CHALLENGE S system architecture maximizes performance with fast processors in a high- capacity bus. It supports the MIPS R4600, the MIPS R4400TM CPUs and the new MIPS R5000(TM) processor. The system bus provides 267MB/sec bandwidth within the system and to external peripherals.

Designed for distributed computing, the CHALLENGE S server provides outstanding performance in a client/ server environment. Multiple EthernetTM support, or higher bandwidth networking options including FDDI and ATM, ensure optimum network performance. The CHALLENGE family hardware provides industry-leading NFSTM file server performance combined with SGI® WebFORCE TM World Wide Web software, making CHALLENGE S ideal for digital media content on the Web.

Challenge DM

The Challenge DM network resource server is the entry-level system to the symmetric multiprocessing family of high end servers. Challenge DM is a throughput powerhouse focusing on the needs of the database, file serving, Web serving, digital media, and real-time markets.

The CHALLENGE DM system architecture is powerful and flexible. It is based on a 1.2GB/second sustained bus which eliminates time-wasting contention between the various I/O, memory and compute subsystems. The system supports from one to four MIPS® R4400TM CPUs with 1MB or 4MB of secondary cache each, multiple gigabytes of CPU memory and up to three 320MB/second POWER ChannelTM-2 I/O subsystems.

The CHALLENGE DM network resource server comes standard with one POWER channel-2 subsystem. Each POWER channel-2 contains an Ethernet controller, two 20MB/second FAST/WIDE 16-bit SCSI-II controllers (configurable as single-ended or differential), two 19.2KB RS232 serial ports, a 38.4KB RS422 serial port, a parallel port, and supports up to two optional daughter boards (HIO modules). External storage is provided via the CHALLENGE Vault L, CHALLENGE Vault XL and CHALLENGE RAID. With these products, CHALLENGE DM supports a maximum capacity of more than 3.7TB of on-line, non-RAID storage and 10TB RAID storage, respectively.

The CHALLENGE DM server expands and customizes through industry standard VME-64 based adapters or HIO modules. Among the expansion options: industry-leading performance HiPPI; ATM adapter; 8-port Ethernet adapter; and a dual attach FDDI.

CHALLENGE DM is an ideal performance database platform for serving content-rich information across the Web. It also serves compressed and uncompressed digital video, digital audio, and static imagery. It can read and distribute in real time using striped SCSI disks in communication with a high-speed HiPPI adapter. Compressed digital video can be distributed from a SCSI disk to any number of decompression devices. CHALLENGE DM manages constant bit rate data streams in a deterministic manner. It uses the IRIX real-time operating system in conjunction with the XFS filesystem, which allows a process to reserve bandwidth on a disk.

CHALLENGE DM provides industry-leading file serving perfomance for mid-range NFS file serving. High-performance networking options, optional high availability capability, excellent systems administration tools, and fast and reliable backup and storage tools come together in a complete solution.

CHALLENGE DM is also an ideal server for archive and hierarchical storage management requirements. Near-line storage ranges from a DLT mini-library with 70GB of storage per device to a multiterabyte D1 based tape library. The IRIXPro suite of powerful and flexible system administration tools can easily manage multiple servers.

Challenge L

The CHALLENGE L network resource server is a compute and throughput powerhouse for deskside and data center applications in the database, digital media, integer compute real-time and high-performance file serving markets. CHALLENGE L can handle extraordinary data management tasks, shifting yesterday's compute paradigm from high-cost mainframes to the industry-leading performance of today's cost-effective symmetric multi-processors.

The powerful, flexible CHALLENGE L system architecture is based on a 1.2GB/second-sustained bus, which minimizes time-wasting contention between I/O, memory and compute subsystems. The system supports from two to twelve MIPS® R10000 TM CPUs, multiple gigabytes of CPU memory, and up to six 320MB/second POWER Channel-2TM I/O subsystems.

The CHALLENGE server's flexible hardware and software configurability allows performance optimization, as well as tangible growth paths to meet future computing demands. The server comes standard with one 320MB/second POWER Channel-2 subsystem. Each POWER Channel-2 subsystem contains an EthernetTM controller, two 20MB/second Fast/Wide 16-bit SCSI-II controllers (configurable as single-ended or differential) a 50MB/second VME controller, two 19.2KB RS232 serial ports, a 38.4KB RS422 serial port, a parallel port, and supports up to two optional daughter boards (HIO modules).

External storage is provided via the CHALLENGE Vault L disk expansion chassis. With this product, the CHALLENGE server can support up to 5.6TB of on-line, non-RAID storage and 17.4TB of RAID-5 and RAID-3 storage.

The CHALLENGE L server is ideal for cutting edge data warehousing applications. With CPU and I/O expandability to meet the needs of interactive terabyte databases, this system can be grown to manage the data extraction and data hygiene needs of each customer. HIO-based SCSI expansion provides higher database performance by reducing I/O interrupt overhead. Large secondary cache also increases overall database performance. In addition CHALLENGE uses the IRIX symmetric multi-processing operating system in conjunction with the XFSTM, a native log-based filesystem. This provides a high level of data integrity, quick data recovery, and a robust environment suitable for the parallelism and throughput requirements of data warehousing.

CHALLENGE L serves the realm of extreme data as a deployable machine unmatched in CPU and I/O throughput capabilities. Complementing these hardware strengths are software advances which create industry-leading file server performance for the enterprise. For the most creative graphics workstation users manipulating a range of digital media data, IRIX offers deterministic I/O, large file and partition sizes, and intuitive data management through the IRIXPro suite of flexible GUI-based administration tools. Performance-hungry supercomputer sites requiring powerful front-end servers with direct and buffered I/O capabilities also enjoy efficient utilization of high-performance networks (HiPPI, FDDI) and sparse file support.

ChallengeXL

The CHALLENGE XL network resource server is a compute and throughput powerhouse for deskside and data center applications in the database, digital media, integer compute real-time and high-performance file serving markets. The CHALLENGE XL can handle extraordinary data management tasks, shifting yesterday's compute paradigm from high-cost mainframes to the industry-leading performance of today's cost-effective symmetric multi-processors.

The powerful, flexible CHALLENGE XL system architecture is based on a 1.2GB/second sustained bus, which minimizes time-wasting contention between I/O, memory and compute subsystems. The system supports from two to thirty-six MIPS® R10000 TM CPUs, multiple gigabytes of CPU memory, and up to six 320MB/second POWER Channel-2TM I/O subsystems.

The CHALLENGE server's flexible hardware and software configurability allows performance optimization, as well as tangible growth paths to meet future computing demands. The server comes standard with one 320MB/second POWER Channel-2 subsystem. Each POWER Channel-2 subsystem contains an EthernetTM controller, two 20MB/second Fast/Wide 16-bit SCSI-II controllers (configurable as single-ended or differential) a 50MB/second VME controller, two 19.2KB RS232 serial ports, a 38.4KB RS422 serial port, a parallel port, and supports up to two optional daughter boards (HIO modules).

External storage is provided via the CHALLENGE Vault XL. With this product, the CHALLENGE servers can support up to 5.6TB of on-line, non-RAID storage and 17.4TB of RAID-5 and RAID-3 storage.

The CHALLENGE XL server is ideal for cutting edge data warehousing applications. With CPU and I/O expandability to meet the needs of interactive terabyte databases, this system can be grown to manage the data extraction and data hygiene needs of each customer. HIO-based SCSI expansion provides higher database performance by reducing I/O interrupt overhead. Large secondary cache also increase overall database performance. In addition CHALLENGE uses the IRIX symmetric multi-processing operating system in conjunction with the XFSTM, a native log-based filesystem. This provides a high level of data integrity, quick data recovery, and a robust environment suitable for the parallelism and throughput requirements of data warehousing.

The CHALLENGE XL serves the realm of extreme data as a deployable machine unmatched in CPU and I/O throughput capabilities. Complementing these hardware strengths are software advances, which create industry-leading file server performance for the enterprise. For the most creative graphics workstation users manipulating a range of digital media data, IRIX offers deterministic I/O, large file and partition sizes, and intuitive data management through the IRIXPro suite of flexible GUI-based administration tools. Performance-hungry supercomputer sites requiring powerful front-end servers with direct and buffered I/O capabilities also enjoy efficient utilization of high-performance networks (HiPPI, FDDI) and sparse file support.

Origin200

Origin 200 delivers optimized solutions for media distribution, interoperable file serving, multitiered database management, performance-intensive computing, and Internet applications. With UNIX® tools and performance, scalability in all dimensions, interoperability services, and an entry price comparable to PC-based servers, Origin 200 makes perfect business sense.

Origin2000

Origin 2000 is available in 2 configurations, Deskside and Rack.

Origin^TM 2000 Deskside

Origin 2000 provides the ability to seamlessly grow with your business. Utilizing an advanced version of ccNUMA architecture, Origin servers provide true scalability in all dimensions, and deliver a unique combination of industry-leading performance, expandability, availability and compatibility. The entry into the product line is a one- to eight-processor deskside system, which can be connected as a module or added in a rack-mounted configuration as your business computing needs demand.

Origin 2000^TM Rack

Origin 2000 is truly powerful and flexible, allowing you to expand, repurpose, or redeploy your hardware investment as your business demands change. Rack-mounting and the unique CrayLink^TM Interconnect allow you to add modules and move from one to 256 processors in a single-image system, providing up to 512GB of shared physical memory and 80GB per second of sustained I/O bandwidth.

SGI 1400

Based upon Intel® Pentium® III XeonTM processors, the SGITM 1400 servers combine SGI's commitment to high-performance scalable servers with industry-standard technologies, enabling superior solutions at affordable prices. SGI 1400 servers enable many new solutions, including solutions for Internet service providers, education and research institutions, manufacturing firms, entertainment and digital media providers, and software developers.

The SGI^TM 1400L and SGI^TM 1400M four-way servers upport SGI Linux and Windows NT respectively.

SGI 2100

The SGI 2100 server leverages the success of the award-winning Origin^TM servers from SGI and delivers enterprise-caliber performance to the volume server market at an aggressive entry-level price point. SGI 2100 offers big data capacity, interoperability, scalability from two to eight processors, high availability, and the industry's most powerful server architecture, ccNUMA (cache-coherent non-uniform memory access) while delivering high bandwidth for I/O-intensive applications.  It is the ideal server for data-intensive tasks and budget sensitive analytical computing.

The SGI 2100 server brings a price/performance advantage to high-volume applications such as Web serving, file serving and media streaming, in addition to supporting compute-intensive solutions in the realms of technical/scientific computation, bioinformatics, and video serving.

Visual Computing

Visual Workstations:

Silicon Graphics® high-performance visual workstations can be scaled for any combination of power, interactivity, and price point.

WindowsNT® Intel Systems

Silicon Graphics 320

The Silicon Graphics 320 visual workstation delivers astonishing graphics and media capabilities at a surprisingly affordable price. With dual-capable Intel® Pentium® II or Pentium III processors, 100% Windows NT® application compatibility, and the advanced Cobalt^TM graphics chipset providing 2D and 3D graphics acceleration and multimedia capabilities, Silicon Graphics 320 offers the best price/performance in its class--putting unprecedented power in the hands of visual computing professionals everywhere.

Silicon Graphics 540

The Silicon Graphics 540 visual workstation is the ultimate Windows NT graphics and media workstation, delivering unequaled performance and quality for the most demanding applications. With its incomparable 2D and 3D graphics, professional digital video and multimedia capabilities, Intel Pentium III Xeon^TM four-processor scalability, and extensive expansion capacity, Silicon Graphics 540 represents a breakthrough in computing.

UNIX® Systems

Indy

INDY systems are an ideal platform for mechanical engineering and design, media authoring, color publishing, film and video, software development, education and world wide web authoring and serving. INDY systems can also be ordered with Modeler, WebFORCE or Studio software -- solutions that can help you get the job done.

The MIPS RISC-based Indy delivers exceptionally high floating-point and integer performance because it integrates the CPU, FPU, and cache on-chip, eliminating the delays that result from passing signals between chips.

Indy supports several MIPS® RISC processors, including the affordable R4600 ) and the powerful R4400. The simple modular design supports easy CPU upgrades, so you can always take advantage of the latest technology.

The 64-bit Indy architecture contains multiple high-speed buses which efficiently move information through the system, maximizing the benefit of its memory, CPU, and graphics and I/O subsystems, and running your applications faster.

Indy incorporates the sights and sounds of the real world into your work. Indy integrates CD-quality audio and built-in video-in, and includes a full-color digital video camera and microphone plus a rich assortment of tools and applications to capture and create compelling content. With Indy you can select the most effective medium for capturing an idea - a 3D model, movie, image, or sound bite.

The Indy 8-bit or 24-bit XL color graphics subsystem is ideal for 3D as well as 2D imaging. For phenomenal visual computing, the Indy XZ graphics option uses SGI Geometry Engine® hardware graphics accelerators to manipulate complex 3D models.

Indigo2

The Indigo2 workstations are the systems of choice to compute large databases, render complex models and images, and solve the most challenging technical and creative computing problems.

Indigo2 systems are used extensively in key industries including manufacturing, entertainment, visual simulation, chemistry, oil and gas, geographic information systems (GIS), medical imaging and prepress.

Indigo2 Impact

Introduced in July 1995, the Indigo2 IMPACT line includes three models: Indigo2 Solid IMPACT, for 3D gouraud-shaded solid modeling; Indigo2 High IMPACT, for leading 3D performance combined with advanced features such as accelerated texture mapping and image processing; and Indigo2 Maximum IMPACT, for twice the performance of Indigo2 High IMPACT as well as higher color and screen resolutions. The flagship of the product line, Indigo2 Maximum IMPACT boasts the richest feature set and is well suited for advanced design, styling, digital prototyping and character animation. The Indigo2 Solid IMPACT^TMis an entry graphics workstation. Specifically designed for 3D solid modeling, Indigo2 Solid IMPACT offers four times the overall graphics performance and a significantly lower price than its predecessor, Indigo2TM XZ.

Indigo2 IMPACT^TM 10000, workstations are based on the  MIPS® RISC R10000 TM microprocessor. The Indigo2 IMPACT 10000 workstations are the first desktop systems to fuse the computational power of the R10000 processor with the unmatched realism and interactivity of Indigo2 IMPACT graphics in a complete 64-bit software environment. Indigo2 IMPACT 10000 systems achieve two to three times the performance of the MIPS R4400TM processor, previously MIPS' fastest general-purpose processor, on real applications. Indigo2 IMPACT 10000 workstations also exceed the computational performance of POWER Indigo2^TM workstations based on the MIPS R8000TM processor, previously the performance leader for floating-point intensive analysis applications.

On the industry standard SPEC95 benchmark suite, the estimated ratings of the Indigo2 IMPACT 10000 are 8.9 SPECint95 and 12.5 SPECfp95 (8.2 SPECint_base95 and 11.5 SPECfp_base95). These performance ratings are the fastest of any desktop system available or announced from any other vendor, exceeding top desktop systems from Hewlett-Packard, Sun Microsystems, IBM and Digital Equipment Corp.

Silicon Graphics O2

The O2 visual workstation is the affordable, innovative UNIX system that combines SGI performance in a highly integrated, feature-rich user environment. The O2 workstation's unique Unified Memory Architecture (UMA) combined with either the QED RM5200 or the MIPS® RISC processor gives you a high-performance system with integrated 32-bit graphics, image processing, video, and compression features for outstanding application performance. The fully integrated video features offer either analog or digital video for your own desktop production facility. With this combination of performance and outstanding features, O2 delivers outstanding price/performance for UNIX applications.

Designed for the critical path, the O2 visual workstation suits professionals -- the creative and engineering teams that drive a company's competitive edge in the sciences, manufacturing, entertainment, and other demanding industries. O2 systems merge leading compute performance with the industry's first high-performance Unified Memory Architecture (UMA) to deliver incredible internal bandwidth and accelerate graphics and compute-intensive applications.

Interactive Graphics, Video, and Image Processing  The Unified Memory Architecture enables stunning 3D graphics, powerful image processing, and real-time video processing far beyond any other machine available in its class. These features make the O2 visual workstation the ideal platform for scientific visualization, CAD modeling, animation, imaging and visual simulation. Standard compression hardware and bundled digital media software transforms any user into a video editor, while providing the quality video professionals demand.

The Power Path  Transcending other workstation designs, the O2 system delivers extraordinary system throughput and a strong growth path. Unlike other workstations, O2 visual workstations enable users to easily migrate their applications to higher-performance Silicon Graphics systems such as the Octane power desktop and Onyx2 graphics supercomputer. Users can create 3D models or designs on the O2 visual workstation and move them to higher performance systems when the data sets grow or if they require more sophisticated capabilities such as real-time rendering or virtual fly-throughs. Users gain the ultimate power path.

Silicon Graphics Octane

The Octane visual workstation, with its revolutionary system architecture, combines high-performance single or dual MIPS® R12000® processors with the world's most advanced desktop graphics to deliver new levels of application performance, allowing users to simultaneously tackle more complex tasks such as design and analysis or motion modeling and behavior scripting. Octane visual workstations also offer a choice of industry-leading, real-time analog and digital video options that allow you to handle either simple video editing or up to two streams of uncompressed digital video. Octane also has the bandwidth to handle uncompressed HDTV content. With this combination of performance and features, Octane is the true work-horse of the SGI visual workstation product line.

Because of its modularity, you can upgrade Octane as your needs change.  Octane lets you take full advantage of its advanced graphics, CPU, and I/O subsystems by providing a system architecture with low latency and high bandwidth (two-way 1.6GB-per-second peak for each XIO port). The upgradable, scalable bandwidth ensures that Octane can take advantage of future hardware technologies and can keep up with the increasing performance requirements of mainstream and leading-edge application software.

Octane takes graphics performance to the next level with the E-Series graphics. By making extensive use of dedicated processing hardware, Octane optimizes visualizations, clarifying problems and speeding your progress toward solutions. Octane graphics, combined with one or two R12000® compute engines, lets you execute more tasks on your desktop – simultaneously tackle both design and analysis - with system responsiveness that can keep up with your thought process. Coupled with OpenGL®, the industry-standard open graphics library, Octane translates into optimal application performance for power users.

The Octane system's graphics acceleration subsystem includes a hardware Geometry Engine® processor, dedicated rasterization, and the ability to take advantage of a texturing engine if installed. The dedicated frame buffer memory is specifically tuned for handling 3D images and texture caching memory.

Octane/SE  For solid modeling applications, Octane/SE brings high-end desktop graphics performance to mainstream engineers and technical users. The entry-level  Octane/SE system includes a single Geometry Engine processor and can be configured with a texture subsystem or later upgraded to add enhanced realism.

Octane/SSE  Octane/SSE uses two hardware Geometry Engine processors and two raster engines for twice the solid modeling performance of an Octane/SE system. An Octane/SSE system fills two XIO slots, leaving two more slots for high-speed networking and peripheral options. Octane/SSE is the ideal machine for large solid modeling, mechanical analysis pre- and post-production processing, and untextured 3D animation. If your requirements change--a new project or application--hardware texture support can be added at any time.

Octane/MXE  Octane/MXE sets the graphics performance standard with two hardware Geometry Engine processors, two raster engines and the addition of a full-performance texture subsystem. An Octane/MXE system fills two XIO slots, leaving two or more slots for high-speed networking and peripheral options. The right choice for users with demanding visualization needs, Octane/MXE lets you accomplish more on your desktop whether you are working with digital prototypes, virtual reality, or the most complex 3D models.

Silicon Graphics Oynx

With 2 to 24 MIPS® R4400 or R10000 processors, Onyx combines the power of symmetric multiprocessing with high-end InfiniteReality and RealityEngine² graphics.

Deskside (2-4 CPUs) or rack chassis (2-24 CPUs)  1-3 InfiniteReality or RealityEngine² subsystems in a single rack chassis  8-way interleaving memory: 64MB to 16GB  Up to 68.8GB internal disk; expandable to 2TB total storage.

Photo-realistic, real-time image generation; innovative broadcast virtual sets; precise image exploitation; full-size, digital, mechanical prototypes—these demanding visual computing applications push graphics, image, and video-processing requirements to their limits. With an advanced feature set, the SGI® Onyx InfiniteReality visualization supercomputer can concurrently process graphics, imagery, and video data while offering users real-time interactivity.

Advanced technology aircraft trainers, image exploitation systems, on-air virtual sets, and broadcast special effects applications demand solid frame rates. Drop a frame and you're out of the game. InfiniteReality is the only system in the world that provides performance levels to meet these demands and then extends them to other applications, including digital prototyping, medical diagnostic imaging, oil field reservoir simulation, and scientific visualization.

The InfiniteReality visualization pipeline exploits parallelism in all areas and offers an innovative feature set that goes well beyond traditional graphics-processing systems.

·        Bicubic, trilinear, and bilinear image filtering 

·        3D textures for fully interactive volume rendering 

·        Texture/image download rates up to 220MB/sec 

·        Dynamic resolution for guaranteed frame rates 

·        Geospecific texture support for processing realistic geographic scenes 

·        Dynamic characters for creating interactive virtual actors

·        Dedicated image processing support 16-bit monochrome support for high-quality medical and satellite images

·        Support for two display channels in base configuration or up to eight display channels as an option

To facilitate high-quality textured image generation, InfiniteReality is available with 16MB or 64MB of texture memory and supports both trilinear MIP mapping and bicubic texture filtering. For applications that demand the merging of aerial photographs with geographic terrain, geospecific texturing lets you fly continuously across vast land expanses or zoom from satellite views of the world down to the earth's surface.

Silicon Graphics Oynx2

Onyx2 graphics workstations combine the ultimate in supercomputing and visualization technologies and are designed to simultaneously process 3D graphics, imaging, and video data in real time. Configurable modules boast scalable system bandwidth and low-latency memory access while supporting a familiar programming environment. Onyx2 allows you to attain interactive and dynamic rendering and stimulates creativity and profitability through unique features such as clip-mapping, texture paging, volume rendering, anti-aliased full-frame HDTV display, and multichannel, multioutput immersion support.

[ Home ] [ Up ] [ Challenge and Power Challenge SCSI Configuration ] [ Challenge and Onyx Deskside Card Layout ] [ Changing File and Directory Permission ] [ Changing System Date and Time ] [ Changing System Name ] [ Onyx or Challenge Disk Formatting Fails Using Stand Alone fx ] [ Onyx Challenge Mezzanine Card Options ] [ Disk Restoration ] [ Initialization of IRIX Disk Drives ] [ Partitioning Large Hard Disk Drives For Use With IRIX 5.X ] [ Removing Root Password From SGI Systems ] [ SCSI Revision Levels ] [ Setting Default Printers on SGI Systems ] [ Setting The Network Address on SGI Computer ] [ SGI System Maintenace Menu ] [ SGI Systems Descriptions and Uses ] [ Clone SGI Hard Disk ] [ PROM Password Removal ]

Send mail to jokeefe@hummingbirdcomp.com with questions or comments about this web site.
Copyright © 2007 Hummingbird Computing
Last modified:  Thursday, October 18, 2007