Download PDF
Altair > Case Studies > Weizmann Institute Chemistry Faculty Enhances Cluster Performance with PBS Professional
Altair Logo

Weizmann Institute Chemistry Faculty Enhances Cluster Performance with PBS Professional

Technology Category
  • Infrastructure as a Service (IaaS) - Cloud Computing
  • Sensors - Level Sensors
Applicable Industries
  • Chemicals
  • Life Sciences
Applicable Functions
  • Facility Management
  • Product Research & Development
Use Cases
  • Inventory Management
  • Usage-Based Insurance
Services
  • System Integration
The Challenge
The Weizmann Institute of Science's Chemistry Faculty faced a significant challenge in managing their high-performance computing (HPC) facility. The HPC cluster, consisting of 1242 cores, served hundreds of faculty members and was in the process of expanding to 3096 cores. The cluster was used for a wide range of research fields, including quantum mechanics, protein folding, DNA recognition, turbulence physics, and climate modeling. The installed software was a mix of advanced C and FORTRAN compilers, mathematical libraries, and a variety of free, academic, commercial, and homegrown dedicated software tools. The workload types varied widely, and different teams had different priorities and needs. The computing environment was a complex system of users, resources, requirements, and policies that needed to be carefully managed. The Institute sought a workload management software vendor that could provide consistently high performance, support for complexity in user priorities and profiles, a proactive, collaborative approach to solution delivery, and reliable user support.
About The Customer
The Weizmann Institute of Science is a world-leading multidisciplinary research institution. Hundreds of scientists, laboratory technicians, and research students embark daily on fascinating journeys into the unknown, seeking to improve our understanding of nature and our place within it. Groundbreaking medical and technological applications have emerged from basic research conducted by Weizmann Institute scientists, including amniocentesis, sophisticated laser systems for high-precision diamond cutting, living polymerization, and ribosome structure analysis, for which the Institute’s Professor Ada Yonath was awarded a Nobel Prize in Chemistry. The Institute's Chemistry Faculty uses a high-performance computing (HPC) facility to perform cutting-edge research in a variety of fields.
The Solution
The Institute chose Altair's PBS Professional as their workload management software, replacing their previous solution, Moab from Adaptive Computing. PBS Professional was selected for its flexibility in handling complex requirements and the company's excellent level of service and support. The software was able to manage the complex user policies at the Institute, where both hard and soft limits needed to be managed. The installation process was smooth, taking about 30 minutes, and Altair’s support expert was able to translate policies used with the previous scheduler directly into PBS Professional in less than a week, simplifying the transition process. Altair also provided a custom extension that deletes or resubmits suspended jobs based on certain user choices, ensuring the system load is well managed.
Operational Impact
  • The implementation of PBS Professional has significantly improved the usability and productivity of the Weizmann Institute's Chemistry Faculty. Users have benefited from higher cluster utilization rates, reduced wait times for computing results, and improved productivity due to greater resource availability. The user experience has also been enhanced thanks to a reliable support team. Altair's support engineer provided a custom extension that deletes or resubmits suspended jobs based on user choices, ensuring the system load is well managed. The Institute is now investigating additional Altair features and products from the PBS Works suite. They are also currently expanding the cluster from 1242 cores to 3096 cores, all of which will be managed by PBS Professional.
Quantitative Benefit
  • 30% higher cluster utilization rates
  • Reduced wait times for computing results
  • Improved user productivity via greater resource availability

Related Case Studies.

Contact us

Let's talk!

* Required
* Required
* Required
* Invalid email address
By submitting this form, you agree that IoT ONE may contact you with insights and marketing messaging.
No thanks, I don't want to receive any marketing emails from IoT ONE.
Submit

Thank you for your message!
We will contact you soon.