Enable unprecedented levels of automation and agility with cloud computing solutions.
HPE SGI 8600 System
The world’s most advanced production supercomputer1 – architected for best performance, scale, and efficiency.
Clustered, high performance computing (HPC) system
The HPE SGI 8600 is a scalable, high-density clustered supercomputer system. Its liquid cooling translates into high power efficiency with substantial savings in energy usage.2 With industry-standard processors, interconnect, and software, SGI 8600 offers maximum flexibility and performance. Select from multiple interconnect and server tray options. This unique architecture is coupled with comprehensive system management and services.
Advanced interconnect technologies
HPE SGI 8600 consistently delivers high performance by integrating leading industry-standard interconnect technologies into clustered solutions. Add to this industry-standard software support and close working relationships with interconnect vendors.
Performance with innovative engineering
Building on the proven SGI ICE XA architecture, we leverage a collection of engineering innovations to enable the HPE SGI 8600 to deliver the best possible performance for an industry-standard clustered High Performance Computing (HPC) system.3 These include the most powerful CPUs, fastest memory, latest GPU technologies, highest speed interconnect topologies and tuning of the Message Passing Interface (MPI). Designed to scale to thousands of nodes, the HPE SGI 8600 also is well optimised for deep learning through NVIDIA SXM2 compute trays.
HPE SGI 8600 delivers the best possible performance for industry-standard clustered HPC systems. It supports the highest performing processor SKUs and interconnect technologies and offers world record performance for major message passing benchmarks.4
Density, scale, and efficiency
Innovative cooling and interconnect technologies enable breakthrough density, scale, and efficiency for the HPE SGI 8600. Liquid cooling provides 100% heat removal and substantial savings versus comparably sized air-cooled solutions.
Ease of use
Flexible, easy-to-manage HPC clusters flow from off-the-shelf Linux support and applications as well as HPE system manageability software. Solve complex computing challenges with the HPE software ecosystem. Provision thousands of nodes in minutes.4
Integrated enclosure architecture
The HPE SGI 8600 architecture provides flexible system configuration. A 10.5U compute tray enclosure provides power, cooling, system control, and network fabric for up to 9 compute trays via an integrated midplane.
Operating system choices
HPE SGI 8600 system runs standard SUSE® Linux Enterprise Server, Red Hat® Enterprise Linux®, and CentOS. Different Linux operating systems can be provisioned on different nodes.
Multiple interconnect fabrics
The system supports both InfiniBand and Intel® Omni-Path interconnect technologies with complete flexibility in topology. Choose from All-to-All, Fat Tree, Hypercube, or Enhanced Hypercube fabrics.
The National Center for Atmospheric Research tripled its capacity to run large-scale 3-D models of the atmosphere with its new supercomputer, while reducing infrastructure footprint by 66%.
Server tray portfolio
A quad node Intel® Xeon® Scalable Processor Family compute tray with 2 CPU sockets per node (total 8 CPU sockets). This model delivers maximum Intel® Xeon® Processor Scalable family compute density.
A single two CPU socket node Intel® Xeon® Scalable Processor Family compute tray with support for up to four NVIDIA Tesla SXM2 GPUs with NVLink.
A quad node Intel® Xeon® Phi processor compute tray with one CPU socket per node. All trays utilise an innovative cold sink technology to provide high node density at proper cooling levels
- Compact E-Cell design with up to 36 blades and 144 nodes per rack
- 288 Intel® Xeon® processor sockets per rack
- Integrated dual plane switching, power and cooling
- Direct attached liquid cooling
- 100% heat removal via data centre water
- No heat transfer to data centre room
- Scaling without use of any external switches
- Multiple installations scaling to thousands of nodes
Earth’s greenest supercomputer
The HPE SGI 8600 ranked number one in the June 2017 Green500 rankings, acknowledging it as the world’s most energy-efficient supercomputer.5
Let’s talk about HPC
Speak with an HPE HPC expert and learn how we can help you realise your unique HPC goals with the HPE SGI 8600 and its maximum density, performance, and energy efficiency.
How the National Centre for Atmospheric Research leverages the HPE SGI 8600 Supercomputer to improve weather research.
HPE SGI 8600 Data Sheet
HPE SGI 8600 Quick Specs
What Is Supercomputing
As a leader in the HPC market, Hewlett Packard Enterprise provides unique capabilities for driving innovation into the future. Learn how HPE is approaching the many challenges on the path to Exascale – the future of HPC – the next generation of computing. Register and download the Technical White Paper.
Top500 – the list of the world’s most powerful supercomputers
HPE and Intel® team up to advance customer innovation and expand HPC accessibility
1. Based on a comparison of industry offerings with regard to footprint density, power efficiency, range of tray offerings, support for highest wattage CPUs and GPUs, industry standard OS, integrated networking, power and cooling, and system management software.
2. Substantial savings in cooling costs with Closed-Loop Airflow” which ensures no air within the cell is mixed with data centre air Substantiation: 5 HPE Internal Engineering study “Liquid Cooling Value Proposition”, WWAS February 2017.
3. HPC Wire, “SGI, ANSYS Set New Record for Scaling Commercial CAE Code,” September 2016.
4. HPE Internal testing on SGI ICE XA hierarchical cluster auto-provisioning up to 10K+ nodes, June, 2015.
5. Top500.org’s Green500, June 2017.