SGI

MAID Platforms, SGI UV, Cloucrack C2, GI Rackable & Etc.

MAID Platforms

COPAN™ 400M Native MAID

Ease of Integration with Disk-based Persistent Data-Aware Applications

Today, the vast majority of information is created digitally and is overloading the ability of IT storage infrastructures to cost effectively keep pace with the overwhelming quantity of digital information.

Effectively managing the growth and costs surrounding the massive volumes of data is essential to IT success. Finding ways to reduce costs, simplify management and ease integration are the hallmarks of IT data management strategies.

COPAN 400M provides the keys to solving the Persistent Data challenges—effectively managing the growth, costs and complexities for growing volumes of long-term fixed or static data. By providing easy integration with existing data protection, data archiving, storage tiering, and other data-aware applications. COPAN 400M eases management, reduces costs, and integrates with your existing applications. It provides all of the benefits of Enterprise MAID: unmatched density, scalability, and energy efficiency.

 

 

SGI UV: The Big Brain Computer

  • Focus on Solving Your Problems, Not IT Problems
  • No Limit Computing, Built on Industry Standards
  • World’s Largest In-Memory System for Data-Intensive Problems

SGI UV is the leading scalable shared memory architecture in the industry today and provides compelling advantages in performance and economy by leveraging SGI technology to new levels of scalability and price performance. SGI UV scalable, coherent shared memory system, addresses a broad range of the most demanding data-intensive applications. It enables users to optimally size their systems, achieving the correct balance of compute, memory and storage capability. SGI UV 2 product family can scale a single system image (SSI) to a maximum of 2,048 cores (4,096 threads) because of its innovative NUMAflex®, blade-based architecture. SGI UV 2 includes the latest Intel® Xeon® processor E5-4600 product family, and can operate unmodified versions of Linux such as SUSE® Linux Enterprise Server and Red Hat® Enterprise Linux®. SGI UV 2 is the only solution that currently leverages the power of Intel’s latest CPU beyond 4 sockets and 1.5TB memory.

World’s Largest In-Memory System for Data-Intensive Problems

SGI UV allows users to solve their most demanding problems with fast, scalable performance, operating at up to 64 terabytes of memory and with up to 4096 threads. SGI UV is the ideal platform to accelerate the pace of innovation in all areas of decision support, genomics and bioscience, chemistry and materials, physics, integrative systems science, national security, product design, and other data-intensive fields.

Focus on Solving Your Problems, Not IT Problems

SGI UV operates just like a workstation and is far less complex to manage than traditional scale-out systems with many nodes, and applications can scale without the complexity of multi-instance software. Because of the huge capacity available on SGI UV, users can consolidate complete workflows in a single system, with the lowest IT burden per compute core versus comparable clusters or scale-out systems.

No Limit Computing, Built on Industry Standards

With the ability to start as small as just four sockets, and adding the hardware as needed, users of SGI UV can leave the node memory limits of scale-out computing behind, on an open platform built on industry standard hardware and software. SGI UV 2 is built with the latest Intel® Xeon® processor E5-4600 product family and runs unmodified, off-the-shelf Linux® software. SGI UV also supports scalable graphics accelerator cards, including NVIDIA® Quadro®, NVIDIA® Tesla® K20 GPU computing accelerator and Intel® Xeon Phi™.

 

 

SGI ICE Clusters

SGI ICE X

Petascale-Class, With Clear Roadmap to Exascale Performance

SGI® ICE™ X is the next generation of the world’s fastest distributed memory platform for over three years running. This performance leadership is proven time and again not just in the lab, but at customer sites including the largest and fastest pure compute InfiniBand cluster in the world.

The system combines the powerful Intel® Xeon® processor E5-2600 series platform with a unique board and interconnect design. Running on standard Linux®, SGI ICE X delivers up to 2,304 processor cores — over 53 teraflops of compute power — per rack, easily scaling from 36 to tens of thousands of nodes to address the most challenging compute problems.

SGI ICE X provides nearly a 5x performance density improvement over the previous industry-leading generation, making your applications fly. With its innovative design, a nearly 2.5x boost in density is also assured, saving valuable space in your data center.

SGI ICE X is designed to minimize system overhead and communication bottlenecks that can rob efficiency and scalability. For example, the system offers the highest performance and most scalable system for CFD, now optimized for SGI OpenFOAM®.

Installs Production-Ready in Hours or Days, Not Weeks or Months

Power up and go, at scale, to solve even the most data-intensive problems.

Entirely built on industry-standard hardware and software components, SGI ICE X enables access to the full spectrum of the Linux ecosystem. SGI Performance Suite optimizes the performance of Linux® applications. And reliability, availability and serviceability are ensured with SGI Management Suite, enabling ease of administration including power management, which is all-important in today’s green environment.

Ultimate Flexibility and Seamless Scalability

SGI ICE X is the only system in its class offering expandability within and across technology generations while maintaining uninterrupted production workflow, and is the only platform capable of seamless scalability from tens of teraflops to tens of petaflops. User choice is available in topology and interconnect, power, cooling, CPUs and memory. The integrated bladed design offers rack-level redundant power, as well as cooling via air, cold water and even warm water, for enhanced reliability and availability. The result is a system with unmatched efficiency, performance and overall value.

Also available with storage and visualization options, SGI ICE X sets a new standard for flexibility, simplicity and ease-of-use in scale-out computing. Along with industry-leading professional services and support, SGI enables customers to efficiently overcome the challenges of complex data intensive workflows, leading to accelerated results.

 

 

Cloucrack C2

Cool, Clean and Efficient

CloudRack C2 delivers extreme densities with break through energy efficiency for cluster computing in a unified cabinet design where power and cooling is centralized at the rack level. Its Eco-Logical™ design uses fan-less and cover-less 1U server trays, assuring maximum power usage, cooling efficiency, and staggering server densities with up to 1824 cores per cabinet.

Ideal for Scale-Out Deployments
CloudRack C2 is the ideal choice for data centers looking to maximize server density while cutting operational costs. By leveraging three-phase input power, CloudRack C2 delivers dramatic bottom-line savings by eliminating stranded power (a data center’s power capacity which is paid for, but ultimately unused)

Available in both 24U and 42U sizes, the 24-inch wide cabinets were designed for hot-aisle, cold-aisle data centers.

 

GI Rackable

Half-Depth


1000 Series

2000 Series
1U Rackable servers feature novel half-depth form factor to enable back-to-back mounting. Highest density witH up to 92 servers with 184 processors and 2,944 cores in a single Rackable rack. 2U Rackable servers feature all the benefits of back-to-back mounting with more configuration flexibility. Up to 46 servers can be hosted in a single Rackable rack.

Datasheets for Rackable Half-Depth Servers Based on:

 

Standard-Depth


1100 Series

2100 Series
1U standard-depth rackmount form factor. Single and dual node configurations deliver high density in standard 19″ rack environments. 2U standard-depth rackmount form factor. Single and dual node configurations are suitable for Enterprise workloads and are also ideal for use as HPC cluster head nodes.

SGI Rackable standard-depth, rackmount servers and clusters deliver top value and performance, based on the winning combination of the latest Intel® or AMD® processor architectures and SGI expertise in designing and delivering the most advanced performance computing systems available. Rackable servers support up to 512GB of memory per node in an ultra-dense architecture with up to 32 cores per 1U. In addition, there is support for up to 56 Gb/sec. FDR and QDR InfiniBand, eight-core processors and DDR3 memory. Aggregated together, SGI can deliver some of the most powerful cluster solutions available.

Datasheets for Rackable Standard-Depth Servers Based on:

 

 

SGI® Rackable® Four-Way Servers

SGI four-processor servers offer impressive levels of performance, density and capacity. These systems are ideal for both enterprise and high performance computing environments. Up to 21 H2106 servers can be mounted in a standard 42U cabinet-each with up to 64 AMD Opteron™ 6200 processing cores-for up to 1344 cores per rack. SGI UV 20 is also available, delivering up to four Intel® Xeon® E5-4600 series processors and up to 1.5TB of memory per server

Model Number H2106-G7 H4002-G6
Chassis Profile 2U standard-depth 4U half-depth
Servers/System One quad-socket One quad-socket
Chipset AMD SR5690 + SR5670 + SP5100 AMD SR5690 + SP5100
Max. Processors Four AMD Opteron 6200 or 6300 series processors Four AMD Opteron 6200 series processors
Max. Cores 48 48
Max. Memory 512GB 512GB
Memory Slots 32 32
Memory Type 1600/1333/1066/800 MHz DDR3 ECC reg. 1600/1333/1066/800 MHz DDR3 ECC reg.
Max. Hard Disk Drives & Max. Capacity Six 3.5″ (max. 18TB) 6Gb/s SAS, SATA II or SSD hot-swap drives Two 3.5″ (max. 4TB) 6Gb/s SAS or SATA II hot-swap drives
SATA ports (Onboard) Eight Six
SAS ports (Onboard, Optional) Eight Eight
SAS RAID (Onboard, Optional) JBOD, RAID 0, 1 JBOD, RAID 0, 1
SAS RAID (RAID card, Optional) JBOD, RAID 0, 1, 5, 6, 10 JBOD, RAID 0, 1
Expansion Slots Two PCI-E 2.0 x16 and two PCI-E 2.0 x8
   Slot 6 N/A N/A
   Slot 5 N/A N/A
   Slot 4 PCI-E 2.0 x 16 PCI-E 2.0 x16 or PCI-E 2.0 x8**
   Slot 3 PCI-E 2.0 x 8 PCI-E 2.0 x8**
   Slot 2 PCI-E 2.0 x 16 PCI-E 2.0 x8
   Slot 1 PCI-E 2.0 x 8 PCI-E 2.0 x8
Networking (Onboard) Dual GigE (Intel 82576) Three GigE ( Intel 82574L and Intel 82576EB)
IPMI Remote Management (Optional) IPMI 2.0 + Keyboard, Video and Mouse (KVM) IPMI 2.0 + Keyboard, Video and Mouse (KVM)
Roamer™ Remote Management Card/LCD (Optional) N/A N/A
Video (Onboard) Matrox G200 AST2050
Power Supply 1400W 1+1 redundant auto-switching 180-240 VAC (50-60Hz) 1+1 redundant auto-switching 100-240 VAC (50-60 Hz)
Chassis Mount Standard 19″ rack compatible rail mount Rackable cabinets with back-to-back mounting for double density; standard 19″ rack compatible
Dimensions (HxWxD) 3.5″ (8.9cm) x 17.2″ (43.7cm) x 27.75″ (70.5cm) 7.0″ x 17.6″ x 15.5″

** Slot 4 supports PCI-E 2.0 x16 if jumper JP7 is open, making slot 3 unavailable. If JP7 is closed, slot 4 supports PCI-E 2.0 x8 and slot 3 supports PCI-E 2.0 x8

 

Add comment