It appears that JavaScript is disabled in your web browser. Please enable it to have full system functionality available.

185

RSC Tornado Cluster Architecture '2011

Energy-efficient and easy expandable ultra high-dense RSC Tornado Cluster Architecture combines all advantages of the latest RSC technologies. This architecture enables record computing density for HPC industry – over 269 TFLOPS per cabinet or 128 computing nodes per a 42U cabinet taking only 0.64 m2 footprint.

Advantages

  • Record-level density and performance – up to 128 expandable x86-architecture servers with the highest performance models of Intel Xeon e5-2600 server processors with 145W thermal design power or 269 TFLOPS per standard 42U compute cabinet (80х80х200 cm);

    Peak computing power over 169.57 TFLOPS in one cabinet based on Intel x86 architecture with AVX Instruction Set and over 269.38 TFLOPS with Intel Xeon Phi coprocessors (Intel Many Cores architecture). That provides 265 TFLOPS per sq.m. (based on Intel Xeon E5-2600 v3 processors only) and 420 TFLOPS per sq.m. (by using Intel Xeon Phi coprocessors).

  • Record energy efficiency — PUE factor has a value of 1.06, which is a record for the HPC industry;

    Record energy efficiency — PUE (Power Usage Effectiveness) factor (total system power consumption/power consumption of electronic components) has a value of 1.06, which is a record for the HPC industry. It means that no more than 5.7% of power is used for system cooling.

  • Fully independent and expandable compute node/blade server;

    A comprehensive balanced solution combining:

    • Highest performance models of Intel processors (Intel® Xeon® E5-2697/2699 v3) and widely available standard server boards;
    • Intelligent power management with backup power supply;
    • Expandability with PCI Express interfaces and expansion pack diversity;
    • Support of node hot-swap without impact on adjacent nodes.

     

  • Easy expansion of functionality for specific tasks with a wide range of expansion packs – base node concept and various hardware and software expansion modules enable significant enhancement of system functionality without node replacement or significant modification;

    RSC Tornado Expansion pack: An optional functionality expansion solution that can fully meet specific requirements of different customers. It is a fully tested package that can be used right after installation and includes all necessary expansion cards, RSC liquid cooling components, interconnection boards and mounting elements;

    • RSC Tornado HPC-M Expansion Pack focused on high-performance computing with up to two Intel® Xeon Phi™ coprocessors improves computing node performance to 3.74 TFLOPS;
    • RSC Tornado HPC-G Expansion Pack focused on high-performance computing with up to two NVIDIA® Tesla® K40 GPUs;
    • RSC Tornado BigData Expansion Pack provides price-performance optimized configuration for high-performance data processing;
    • RSC Tornado VDI Expansion Pack is intended for virtualization and remote workplace access and for CAD/CAM/CAE fields. It is based on AMD FirePro™ S10000 or NVIDIA GRID™ K1/K2 cards and on Intel® SSD DC P3700 high-performance solid-state drives;
    • RSC Tornado Security&Protection Expansion Pack is developed specifically for customers from financial vertical market and other users that require maximum access speed and data protection;
    • Hardware and software package RSC SDM Expansion Pack implements Software Defined Management concept developed by RSC specialists for RSC Tornado and RSC PetaStream computing nodes improving manageability and accessibility of computing systems for user applications. Flexible combination of management agents (infrastructure components, computing nodes, scheduler and user application libraries) within the scope of this concept enables creation of new Task-Oriented Management Applications (TOMA) that improve computing system efficiency and reduce operating costs. One of the examples of this is an application for adaptive management of task scheduler priorities that considers available power resources and actual capacity of data center cooling systems to prevent task termination and resulting loss of critical data on temporary infrastructure failures;
    • Other specialized expansion packs can also be created on specific requests of the customers. This approach has significantly improved availability of specialized configurations and made modernization process much more easy while preserving record compactness, computing density and power density of solutions based on RSC Tornado cluster architecture.

     

  • Universal chassis cabinet with a record-breaking RSC liquid cooling system with cooling capacity up to 100 kW per cabinet;

    Chassis cabinet contains all elements required for work including: bearing chassis, system for cooling liquid distribution over computing nodes, power supply to computing nodes. Maintenance work can be performed only at one side of the cabinet.

  • Cost (TCO) reduction

    Broadly available components (server boards, memory modules, processors) and record breaking computing efficiency (96% by LINPACK test) enable significant reduction of costs and RSC liquid cooling system eliminates the need to install expensive precision air conditioning systems at the server room.

  • High reliability (no moving components such as fans and traditional hard disks, efficient RSC liquid cooling system).

RSC Tornado Philosophy

RSC Tornado Cluster Architecture

Solutions based on RSC Cluster Architecture

 




Share Page