The Appro Xtreme-X™ is ideal for customers looking for system scalability, high performance and high availability features with robust remote management capabilities in an open supercomputer cluster architecture. These clusters are grouped together to make a fully integrated Scalable Unit (SU) that can be provisioned and managed as a stand-alone supercomputer to deliver sustained capacity computing performance.

New! The Appro Xtreme-X2 system has been upgraded to support the latest AMD Opteron™ 6100 Series Processor with options for 8 to 12 processor cores. The combination of higher core density, new AMD 5690 chipset and PCI Express® Gen 2 enables significant performance advantages, with up to 48 total cores per node providing improved memory bandwidth by 1.3x of previous 2- sockets server generations while delivering choices in energy footprints to match memory and compute intensive workloads and power requirements.

Appro Xtreme-X2 System Highlights:
- Improved power and cooling efficiency
- Support for 8 to 12 Core AMD Opteron™ 6100 Series processors
- Fully integrated, tested solution with a variety of system configurations
- 64 nodes/128 CPUs/1,536 cores
- Peak Performance/Rack: 13.5TF
- Memory Capacity/Rack: 6.0TB
- Option configuration with the latest NVIDIA Tesla 20 Series GPU available

The Xtreme-X2 Supercomputer offers three configuration models:
EL Series – Entry-level system supporting single-rail fat-tree configuration
                      Includes support option for NVIDIA Tesla 20 Series GPU
MR Series - Mid-Range system supporting dual rail fat-tree configuration
HE Series – High-End system supporting dual rail 3D torus configuration
*All Xtreme-X Supercomputers are powered by Appro Cluster Engine Management Software

Appro Reference Architecture Designs
Appro demonstrates performance and scalability using building block and scalable units reference architecture designs based on customer requirements. The Appro Reference Architecture Designs are based on a 42U Standard Rack Cabinet. It enables a supercomputing environment to be built with up to 1000TF while offering a complete stack of management software: network, servers, cluster and storage.

Scalable and Balanced Architecture
The usable performance of a large cluster depends on a balanced architecture and well designed support software. The memory and I/O system must support the performance of the processors or the entire machine will run no faster than the slowest components.

Water Cooled Rear Cabinet Door Options
It provides up to 35kW of cooling using a passive design with no fans, moving parts or electrical connections. It is 100% sensible cooling with no condensation. It installs in minutes with no rearrangement of enclosures required while allowing 8x more compute power versus other type of solutions.

Appro Cluster Engine™ (ACE) Management Software
Appro Cluster Engine™(ACE) Management Software features a complete lights-out management solution. This software suite offers a web-based management interface that is easy-to-use, making it possible to control the Appro Xtreme-X™ from any location. The management modules include Network Management, Server Management, Cluster Management and Storage Management. In addition, ACE supports multiple networking topologies, network booting configuration and failover features to achieve maximum reliability, performance and high availability. It supports root file systems offering instant provisioning for rapid, standard Linux installs on large diskless systems allowing them to boot 64 to 6,400 blades in the same time.

Ideal environment
For demanding applications that require higher reliability and better system management. Delivers a balanced supercomputer cluster building block for a wide range of HPC applications such as electronic design automation, aerospace and automotive engineering simulations, petroleum exploration and production, scientific visualization for oil discovery and recovery, research in seismic, weather and environmental sciences, defense and classified research/production, and financial modeling and risk management.