PCM Clustering Software Debut - Parallel Computing Manager Cluster Software Introduced

Share Article

Dauger Research, Inc., introduces the Parallel Computing Manager (PCM) software for running and and managing clusters in the enterprise. From a long-recognized leader in cluster solution providers, PCM is an unmatched combination of powerful tools to manage and maintain cluster resources, maximizing the power of cluster computing for its corporate and institutional clients.

Dauger Research, Inc., introduces the Parallel Computing Manager (PCM) software for running and and managing clusters in the enterprise. From a long-recognized leader in cluster solution providers, PCM is an unmatched combination of powerful tools to manage and maintain cluster resources, maximizing the power of cluster computing for its corporate and institutional clients. Today PCM debuts as a Universal Application.

"Derived from eight years of experience in clustering, PCM provides administrators the way to control productive, reliable, and powerful clusters", said Dr. Dean Dauger, President of Dauger Research, Inc. "By short-circuiting the aggravation of complex configuration, fragility, and arcane commands, PCM provides institutional customers powerful, scalable tools to best support and manage computational resources for supercomputer-compatible clustering."

Using IEEE award-winning, industry-leading technologies, PCM debuts with features like those at supercomputing centers. IT administrators can use PCM to assign users quotas and password-based cluster access. PCM technology has users share cluster resources by configuring user accounts, groups, and users' access to computational power, all configurable in a straightforward, easy-to-use, graphical user interface. PCM even utilizes Apple's CDSA implementation for secure data encryption of its the user database as well as during launch and forwarding of parallel jobs. PCM also features a client version, from which users can only launch jobs into the cluster.

PCM explicitly supports clusters configured with a head node protecting compute nodes as a firewall and NAT router in a Linux-style network topology. The software can launch jobs onto such a cluster, check its status, and retrieve jobs via the head node. PCM's multiprocessing support enables users to customize the "virtual node" count for any individual node, including Intel Core's. PCM supports a Grid job type for distributed launching of many single-processor executables with point and click, all execution environments (Carbon, Cocoa, Mach-O, Classic, Unix script, AppleScript, and Universal) available on current Apple hardware and supports five distinct implementations of the Message-Passing Interface (MPI) industry standard. Finally, PCM uses Bonjour to ease node cluster configuration.

The Dauger Research Vault presents eight tutorials extensively describing how to develop parallel applications and algorithms using the MPI standard. Visitors to the web site can download GUI and command-line installers containing a trial PCM, sample source code, a Software Development Kit, and full documentation.

PCM is available today for US$250 for the first compute node then US$200 for each node thereafter. Customers may order PCM using the forms on our web site or online through the Dauger Research Store.

PCM requires networked Macintoshes and/or Xserves running Mac OS X 10.3.9 or later and/or Mac OS X Server 10.3.9 or later with 4 MB of available RAM and 2 MB of disk space.

Profiled and honored on national television by the William Shatner-hosted "Keeping America Strong" show, Dauger Research, Inc. makes high-performance computation and visualization easy to use and accessible to users. Our award-winning team, to better accomplish our scientific goals, reinvented the cluster computer in 1998, pioneering easy-to-use, high-performance clusters. Dauger Research, Inc., is committed to bridging people to the scientifically and technically complex.

###

Share article on social media or email:

View article via:

Pdf Print

Contact Author

Dean Dauger, Ph. D.
Visit website