Altair’s new Data Center GPU Management Tool is now available to Nvidia HPC Customers. With the wide adoption of Graphics Processing Units, customers are addressing vital work in fields including artificial intelligence, deep learning, self-driving cars, and virtual reality now have the ability to improve the speed and reliability of their computations through a new technology collaboration with Altair to integrate PBS Professional.
Features in Altair’s PBS Professional HPC workload management software are now available to Data Center GPU Management Tool NVIDIA DCGM users and enable new functionality and performance levels with:
- Pre-job node risk identification and GPU resource allocation
- Automated monitoring of node health
- Reduced job terminations due to GPU failures
- Increased systems resiliency via intelligent routing decisions
- Increased job throughput via topology optimization
- Optimized job scheduling through GPU load and health monitoring
“We are truly excited to bring this functionality to NVIDIA’s global GPU user community,” said Dr. Bill Nitzberg, CTO for PBS Works, Altair. “They did a great job of enabling the technical collaboration and we look forward to seeing their customers get ever-better computing performance to support their research, development, and analytics applications.”
Duncan Pool, Director Platform Alliances – Accelerated Computing, NVIDIA, notes that, “Altair has been a great partner throughout the process of developing and deploying these new performance tools to our users, aligning well with our intense focus on providing a best-in-class GPU application environment.”
Altair PBS Works technical support staff will work closely with NVIDIA DCGM users running PBS Professional to implement the new functionality on their computing platforms.
Read additional details about how NVIDIA DCGM simplifies cluster administration at the NVIDIA blog.