In this video, Martin Nicholls, Senior Systems Admin.at the University of Queensland describes how PBS Works helps schedule HPC jobs at the university and keeps user happy. Nichols recently presented on “Using PBS Professional Hooks” at the PBS Works User Group meeting 2012 in San Jose. Watch the Presentation or Download the slides (PDF).
In this video, Altair Chairman & CEO James R. Scapa describes the company’s mission to radically change the way organizations design products and make decisions with High Performance Computing.
Recorded at the PBS Works User Group meeting in San Jose on Oct. 2, 2012.
In this video, panelists discuss the next five years in HPC.
Moderator: James R. Scapa, Chairman & CEO, Altair
Panelists:
Recorded at the PBS Works User Group meeting in San Jose on Oct. 2, 2012.
In this video, Sam Goosen, Manager of Product Specialization at Altair presents: Fine Tuning Job Priorities Using the PBS Professional Scheduling Formula.
The PBS Professional scheduling formula will allow administrators to fine-tune the order at which job are scheduled and dispatched to the compute nodes. The administrator can construct a formula based on Python mathematical functions using built-in PBS or custom numerical values. The formula expression can also incorporate policies such as eligible_time, fairshare_percentage, job resources requests, and/or administrator criticality. The scheduling formula is a powerful method for dynamically reordering the job priorities at each scheduling cycle.”
Recorded at the PBS Works User Group meeting in San Jose on Oct. 3, 2012.
In this video, Justin Wood, Systems Engineer at SAIC/Fleet Numerical Meteorology and Oceanography Center presents: Supporting Mission Critical Global Weather Modeling.
Fleet Numerical Meteorology and Oceanography Center (FNMOC) provides global and regional weather prediction to the DoD. This presentation will review how we use PBS Professional to schedule and analyze jobs in order to meet daily deadlines for our customers across the globe.”
Recorded at the PBS Works User Group meeting in San Jose on Oct. 3, 2012.
In this video, Greg Matthews, Deputy PBS Administrator at NASA presents: Combining Quick-turnaround and Batch Workloads at Scale.
NASA Advanced Supercomputing (NAS) uses PBS Professional to schedule and manage the workload on Pleiades, an 11,000+ node IB cluster. At this scale the user experience for quick-turnaround jobs candegrade, which led NAS initially to set up two separate PBS servers, each dedicated to a particular workload. Recently we have employed PBS hooks andscheduler modifications to merge these workloads together under one PBS server, delivering sub-1-minute start times for the quick-turnaround workload, and enabling dynamic management of the resources set aside for that workload.”
Recorded at the PBS Works User Group meeting in San Jose on Oct. 3, 2012.
In this video, Ravi Kunju, Managing Director, Strategy and Marketing at Altair presents: Software Asset Optimization.
Software procurement has become a rising cost of doing business. While many organizations closely manage enterprise IT hardware, the need to better control costs associated with software assets has now become a business priority. This presentation will introduce Altair’s “Software Asset Optimization” (SAO) solution for optimizing the utilization of software assets. SAO is a scalable business analytics application providing enterprise-wide visibility to software inventories and deeper insight to support procurement decision-making and right-sizing software investments.
Altair’s SAO baseline solution has been architected to be rapidly deployed in any environment providing on-demand analytics and interactive dashboards for:
- On-demand access to license inventory/usage/trends
- Identification of shelved / underutilized licenses
- Rapid reconciliation of software requests vs capacity
- Quantitative insight to support vendor negotiations
- Improved software capacity planning and forecasting
Recorded at the PBS Works User Group meeting in San Jose on Oct. 3, 2012.
In this video, Dale Dunlap, Director, HPC / Cloud Technology at Altair presents: HyperWorks On-Demand: Altair’s Cloud Vision.
In the past few years “Cloud Computing” has evolved, and is now fundamentally changing the way that many companies manage their IT resources. High Performance Computing (HPC) in the Cloud presents some unique opportunities and challenges. This presentation will discuss Altair’s perspective on this emerging technology, and how we are aligning our products and services to address these challenges.”
Recorded at the PBS Works User Group meeting in San Jose on Oct. 3, 2012.
In this video, Bill Mannel, VP of Product Marketing at SGI presents: Scale-up in your Scale-out.
Mr. Mannel presents SGI’s solutions for HPC, combining scale-out servers offering ease of management, workflow consolidation, and rapid code development, with scale-out clusters, integrated with storage and software.”
Recorded at the PBS Works User Group meeting in San Jose on Oct. 2, 2012.
In this video, Randall Martin, HPC Infrastructure Manager at Clemson University presents: PBS Professional Implementation Experiences at Clemson University.
Clemson University recently migrated its Top500 cluster from Maui/Torque to PBS Professional. In this session, the migration experiences will be detailed. Clemson’s experience with PBS Professional’s on-demand node provisioning feature in conjunction with Dell’s Advanced Infrastructure Manager (AIM) product will also be detailed.”
Recorded at the PBS Works User Group meeting in San Jose on Oct. 2, 2012.
In this video, Martin Nicholls, Senior Systems Administrator at the University of Queensland presents: Using PBS Professional Hooks: Examples and Benefits.
This talk examines practical uses of the PBS Professional hook mechanism to help achieve site requirements, based on our experiences at The University of Queensland Research Computing Centre. The paper includes descriptions of several real-world applications for job submission hooks (qsub_hook) and includes example code to help commence implementation of a generic hook able to handle both job submission and job modification events.”
Recorded at the PBS Works User Group meeting in San Jose on Oct. 2, 2012.
In this video, Jim Glidewell, Senior HPC Analyst at Boeing presents: Boeing, HPC, and PBS Pro — Twelve Years and Counting.
Boeing has been an industry leader in the application of High Performance Computing (HPC) to complex engineering problems. This presentation will provide insights into the value of HPC as a key technology for aircraft design, and how we use PBS Professional as an enabling technology to provide an effective platform for our engineering users. We will review a number of key features of PBS Professional including peering, queue limits, and hooks, as well as local enhancements in the areas of scheduling and monitoring, which together provide the flexibility we need to address an ever-evolving workload.”
Recorded at the PBS Works User Group meeting in San Jose on Oct. 2, 2012.
Join Cray and Altair on October 10, 2012 for a free webinar covering the Cray XE6m and Altair PBS Professional integrated solutions for affordable supercomputing – and hear about a new PBS Professional offer for the XE6m-200.
To register, go to: https://www3.gotomeeting.com/register/992731398
Cray and Altair speakers will address their products for technical enterprise supercomputing systems – designed to provide a lower cost of ownership and serve a range of users in areas like as manufacturing, weather, life sciences, and research communities. Speakers will cover the Cray suite of XE6m HPC products – highlighting a new, limited-time Cray-Altair XE6m-200 offer – and will explain how PBS Professional has been re-architected for Cray systems.
This 45-minute webinar will address:
Presenters
Details
Register now at: https://www3.gotomeeting.com/register/992731398
In this video, Jon Shelley, HPC Software Consultant Lead at Idaho National Lab, presents: Using PBS Hooks to Extend Your Environment and Lessons Learned.
Extending the PBS environment through hooks on our HPC clusters has helped improve our user experience at the Idaho National Lab (INL). Using hooks, we have integrated our HPC resources, and provided data for reporting requirements and useful information to our users. The hooks to be discussed are checking each job request for malformed select statements and providing an alternative resource request message, determining the effective CPUs that each job requested for reporting purposes, and partially integrating Gold allocation manager to track the usage of our systems. We will also discuss issues encountered when integrating Gold resource allocation manager and a node health script.”
Recorded at the PBS Works User Group meeting in San Jose on Oct. 2, 2012.
Earlier this year, Bright Computing introduced their Cloud Bursting capabilities with a lot of fanfare. Now you can see it working for yourself at the upcoming PBS Works User Group meeting, where Solutions Architect Robert Stober will demonstrate how tightly-coupled Bright Cluster Manager and PBS Professional seamlessly extend HPC clusters into the cloud.
Using Bright Cluster Manager, you can burst into public clouds, such as Amazon EC2, with only a few mouse clicks, without the need for expert knowledge of Linux or cloud computing. Every cluster that runs Bright Cluster Manager is automatically cloud-enabled.
At insideHPC, we’ll be on hand to cover Stober’s presentation is at 10:00 AM on Tuesday, October 2nd. Also presenting at the conference is Jim Glidewell from Boeing, a joint Bright-PBS Professional user; experts from Idaho National Lab, Nissan, Clemson University, FNMOC, NASA, and the host, Altair Engineering.
Read the Full Story or Register Now for the PBS User Group conference.
Advertisement
insideHPC.com is a production of insideHPC, LLC. © 2006-2013 Sitemap
