AUGUST 2008

In This Issue

Events
High Productivity HPC
Software Updates and Releases: TORQUE 2.3.3
Did You Know?: Node Allocation Policies
Online Training Sessions
FAQs
Contact Us

Events

Web 2.0

Cluster Resources will be exhibiting at the Web 2.0 Expo in New York City, September 16-19, at the Javits Center.

HPC on Wall Street

Cluster Resources will be exhibiting at High Performance on Wall Street, September 22, in New York City, at the Roosevelt Hotel.

SC08

Cluster Resources will be exhibiting at booth #623 at SC08, November 15-21, in Austin, TX at the Austin Convention Center.


High Productivity HPC

Where: PCPC Direct Headquarters in Houston, TX
When: October 2nd, 11:30am

PCPC Direct is partnering with Microsoft and HP to host High Productivity High Performance Computing on October 2, 2008. This event coincides with the launch of Microsoft Windows HPC Server 2008.

The event will begin with a buffet lunch and a keynote speaker followed by two breakout sessions. Both Cluster Resources and JOA will be presenting and demonstrating on PCPC’s hybrid boot cluster.

Admission is free. To register, send and email with your name and contact information toevents@pcpcdirect.com


Software Updates and Releases

TORQUE 2.3.3 has been released. To download the latest version, go here

To see changelog details, go here

This latest version of TORQUE was primarily released to correct a minor bug where the pbs_mom would occasionally not connect properly with the pbs_server after network failures. Downloading the latest version of TORQUE is the easiest fix; however, for those who cannot upgrade yet, restarting the pbs_mom should bring the compute node back to full health.


Did You Know: Node Allocation Policies

Node allocation policies allow a site to specify how available resources should be allocated to each job. The algorithm used is specified by the parameter NODEALLOCATIONPOLICY. There are multiple node allocation policies to choose from allowing selection based on reservation constraints, node configuration, resource usage, preferences, and other factors. You can specify these policies with a system-wide default value or on a per-job basis. Please note thatLASTAVAILABLE is the default policy.

Other available algorithms include FIRSTAVAILABLELASTAVAILABLEPRIORITYCPULOAD,MINRESOURCECONTIGUOUSMAXBALANCEFASTEST, and LOCAL.

For more information on Node Allocation Policies go here


Online Training Sessions

Cluster Resources provides live interactive demonstrations to show how easy it is to evaluate Moab. These demonstrations go into detail about the tools we provide that make the evaluation simple and effective.

Part 1 of the training covers the use of Moab in a cluster environment, as well as the following areas: Moab architecture, installation, diagnostics and troubleshooting, admin tools, end user tools, and testing modes.

Part 2 of the training details Moab in a grid environment, including information on grid autonomy, credential remapping and other grid specific capabilities.

Part 1: Clusters

U.S Time Zones
–Wed., Sept. 17, 2008 1:00 PM – 2:00 PM EDT

Part 2: Grid

U.S Time Zones
–Wed., Sept. 24, 2008 1:00 PM – 2:00 PM EDT

If these times do not work for you, please contact us at eval-support@clusterresources.com to schedule a demonstration outside the listed times. EMEA training sessions are added periodically.

To register for training sessions go here


Frequently Asked Questions

Cluster Resources has a Moab Evaluation Product FAQ page located here.


Contact Us

Please send your comments and questions about products, product feedback, tech support, consulting, or custom development to info@adaptivecomputing.com or call us at +1 801-717-3700.


Return to Newsletters Menu

 

© Copyright 2010, Adaptive Computing Enterprises Inc., Provo, Utah. All rights reserved. Moab, Moab Adaptive Computing Suite, Moab Adaptive HPC Suite, Moab Viewpoint, and Adaptive Computing are trademarks or registered trademarks of Adaptive Computing Enterprises Inc. Other marks are the property of their respective owners.