More Xtreme-X Supercomputer Configurations Offered

Appro offers Xtreme-X Supercomputer configuration options for entry level up to high end; also improves cluster management software.

Appro offers Xtreme-X Supercomputer configuration options for entry level up to high end; also improves cluster management software.

By DE Editors

 
More Xtreme-X Supercomputer Configurations Offered

Appro (Milpitas, CA) has announced three new configuration options for its Xtreme-X Supercomputer that address the needs of entry-level,  mid-range, and high-availability users. Additionally, the company reports it has advanced its Appro Cluster Engine (ACE) management software with several new features.

The foundation of Appro Xtreme-X Supercomputer line is a scalable architecture that enables you to group high-performance servers together into a unified, fully integrated Scalable Unit (SU) that can be provisioned and managed as if it were a stand-alone supercomputer. The Appro Xtreme-X Supercomputer, which is based on the 6-core Intel Xeon Processor 5600 Series, is available with interconnect fabric options for both single-rail InfiniBand Network or dual-rail configurations. Standard features include power and cooling designed for maximum efficiency; support for up to 64 nodes, 128 CPUs, and 768 cores; memory capacity of up to 6TB per rack; and a per rack peak performance of 9TF.

The EL Series, the MR Series, and the HE Series are three new configuration options for the Xtreme-X1 Supercomputer. The EL Series serves as an entry-level system with a single-rail fat-tree configuration that, says the company, is designed to meet price/performance customer requirements. Additionally, support for NVIDIA Tesla 20 Series GPUs is available for the EL Series.

The mid-range MR Series supports dual rail fat-tree configuration that is suitable for high-performance, high-availability customer applications, according to Appro. The high-end HE Series system supports a dual-rail 3D torus configuration as well as offering extended features for high-performance and high availability. The company reports that additional features are to be announced available in Q4 2010.

 
More Xtreme-X Supercomputer Configurations Offered

Appro ACE Management software provides Xtreme-X Supercomputers with automatic discovery and monitoring of all network and server hardware, enabling plug-and-play functionality for new equipment. ACE’s diskless booting of compute servers is said to allow 10 to 10,000 compute nodes to boot quickly and simultaneously; however, ACE also supports either local or remote disks, which provides users an option for faster distributed disk access. Individual clusters can be independently configured and provisioned to enable workload and energy management, and each application can locate an optimized cluster environment, freeing administrators to manage the global compute environment and energy consumption.

ACE also provides ongoing optimization for expanding physical and virtual Linux environments, according to Appro. ACE offers users tools to manage the software stack, including workload manager, resource management tools, remote control, and advanced power management functions. Additional features include identification network and server hardware problems and automatic job restarts after a compute node failure. As well, ACE provides a redundant hierarchical system management with all system configuration data stored on the fault tolerant database and file system on the management servers.

According to Appro, its ACE Management Software has been enhanced with such features as automatic assignment of cluster host names and IP addresses as well as automatic management of host database (/etc/hosts). ACE now offers support for full installation for Red Hat or CentOS distribution,  host configuration synchronization, and support for efficient global and cluster-level configuration files. Support and configuration for Grid Engine for batch queuing has been integrated. Miscellaneous new features include fully stateless compute nodes and dynamic instant provisioning; GPU monitoring management support capability; and a revision control system enabling multiple cluster software revisions and versions.

Appro Xtreme-X1 Supercomputer and ACE Management Software are Intel Cluster Ready Certified. For complete details, visit Appro.

Go here for more information on Appro Xtreme-X Supercomputers.

Go here for more information on Appro ACE Management Software.

Access a video on Appro ACE Management Software.

Read more about the Xtreme-X’s and ACE Management Software’s Intel Cluster Ready certifications.

See why DE’s Editor selected Xtreme-X Supercomputers and ACE Management Software as their Pick of the Week.

Share This Article

Subscribe to our FREE magazine, FREE email newsletters or both!

Join over 90,000 engineering professionals who get fresh engineering news as soon as it is published.


About the Author

DE Editors's avatar
DE Editors

DE’s editors contribute news and new product announcements to Digital Engineering.
Press releases may be sent to them via [email protected].

Follow DE
#5121