The flagship product in Altair’s award-winning PBS Works suite, PBS Professional is an EAL3+ security-certified HPC workload management product proven for over 20 years at thousands of global sites. PBS Professional offers powerful, policy-based and topology aware scheduling, million-core scalability, and other capabilities for easily managing any HPC system – from small departmental clusters to the largest, most complex systems on the planet.
PBS Professional makes it possible to easily create intelligent policies to manage distributed, mixed-vendor computing assets, including applications, as a single, unified system. Based on a flexible, policy-driven architecture, PBS Professional continually optimizes how your technical HPC resources are used, ensuring you receive the greatest possible ROI on hardware and software resources. The end result: improved productivity, confidence that you’re meeting service levels, and reduced hardware and software license costs.
Efficiently Distributes Workloads for maximum utilization and ROI
Extensive scheduling options, from peer-to-peer and backfill to fair-share scheduling and formula-based, preemptive scheduling for time and resource critical workloads
Enterprise-wide Resource Sharingprovides transparent job scheduling on any PBS system by any authorized user
• User customizable plugin “hooks” framework to support almost limitless extensibility – hook events include submission, alter, run, pre-execution, job launch, post-execution and periodic node
Formula based scheduling allows you to define any policy and combine existing policies into one formula. It allows you to handle “exceptions” by tuning individual priorities on-the-fly
Why PBS Professional?
Scalability to millions of cores
Flexible plugin framework to meet complex customization requirements
- Intelligent job scheduling -- policy-based, topology-aware -- for maximized system utilization
- Access to product source code
- Standard-setting customer services and support, with experts in 22 countries
For more information: