Download PDF
Altair > Case Studies > Enhancing Research Capabilities with High-Performance Computing: A Case Study of QIMR
Altair Logo

Enhancing Research Capabilities with High-Performance Computing: A Case Study of QIMR

Technology Category
  • Infrastructure as a Service (IaaS) - Cloud Computing
Applicable Industries
  • Healthcare & Hospitals
  • Life Sciences
Applicable Functions
  • Product Research & Development
  • Quality Assurance
Use Cases
  • Inventory Management
  • Visual Quality Detection
Services
  • System Integration
  • Testing & Certification
The Challenge
The Queensland Institute of Medical Research (QIMR), one of Australia's largest and most successful medical research institutes, faced a significant challenge in providing shared High-Performance Computing (HPC) resources to its hundreds of scientists, students, and support staff. The institute, which is home to over 50 separate laboratories supporting six research departments, needed advanced facilities to support its scientists' cutting-edge projects and attract the best researchers. To meet this need, an HPC cluster was established to be shared as a service among the scientific labs at QIMR. However, managing job scheduling and optimizing throughput on this shared resource was a complex task that required a reliable workload management system.
About The Customer
Established in 1945 by the Queensland Government, the Queensland Institute of Medical Research (QIMR) is one of the largest and most successful medical research institutes in Australia. It is recognized worldwide for the quality of its research. QIMR is home to more than 700 scientists, students, and support staff with over 50 separate laboratories supporting six research departments: Biology, Cell and Molecular Biology, Genetics and Computational Biology, Immunology, Population Health, and Joint Programs. Its researchers investigate the genetic and environmental influences of a range of diseases including cancer, asthma, HIV, malaria, endometriosis, and dengue fever.
The Solution
In late 2010, QIMR procured a number of Dell multi-core servers and storage to facilitate the shared HPC cluster. The cluster consisted of a combination of Dell PowerEdge R815 Servers and Dell PowerEdge R710 Servers. After evaluating several commercial and open-source options for a workload management system, QIMR chose PBS Works from Altair. PBS Works is a suite of job scheduling, on-demand computing, and analytics products for managing HPC workload. Its flagship product, PBS Professional, is among the top-rated available products for HPC workload management and job scheduling. In 2011, the team installed PBS Professional and began bringing users on board. In 2012, they moved the HPC cluster into a new purpose-built data center, which provided more potential for expansion, an enhanced operational environment, and more resilient power management.
Operational Impact
  • The implementation of PBS Professional has resulted in significant operational benefits for QIMR. The system has proven to be an easy-to-use yet powerful tool for submitting HPC jobs. Even groups of statisticians who had never used a product like PBS Professional were able to get up to speed very quickly. The Genetics and Computational Biology group, which operates its own Condor cluster, has also started using the central PBS Professional system for processing. System administrators at QIMR appreciate PBS Professional's ease of management from a corporate perspective, and the flexibility and control it offers in managing job scheduling, tasks, and resources. The institute now plans to expand the scale of the cluster as more researchers learn of the availability and benefits of this shared resource.
Quantitative Benefit
  • Support for breakthrough projects
  • Reduced management overhead
  • Higher quality research outcomes

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.