Performance Begins with Univa

Univa® Grid Engine® – Unleash the Power of High-Performance Intel® Omni-Path Fabrics

When it comes to high-performance computing (HPC), simply having the world's fastest processors and interconnects is not enough. To realize maximum value, clusters based on Intel® Omni-Path Fabrics need high-throughput scheduling software that can keep up, wringing every ounce of performance from state-of-the-art HPC environments.

This is where Univa Grid Engine comes in. With over 10,000 deployments, scalability to 200,000 cores, and the industry's broadest support for HPC workloads with 1,000 plus application integrations, Univa Grid Engine is the choice of HPC professionals. Using smarter scheduling, organizations can boost performance, maximize cluster usage, and delight end-users by helping them take full advantage of blazing fast compute clusters built with Intel Fabrics.

Low-latency Omni-Path Fabrics demand low-latency scheduling, and this is where Grid Engine shines. Not only does Grid Engine seamlessly manage large-scale MPI workloads common in CAE, CFD and Oil & Gas, it is equally at home supporting large numbers of smaller jobs as well, scheduling them with high throughput and applying sophisticated sharing policies ensuring that service levels are maintained. With sophisticated placement controls and parallel job management features, Grid Engine can deploy MPI workloads optimally to take full advantage of Omni-Path clusters for improved performance. Grid Engine is also optimized to support a variety of other workloads including those found in Life Sciences, EDA, Financial Modelling, CAE Parametric Sweeps and other simulation activities. With Grid Engine, users of HPC clusters based on Intel Fabric technology benefit from maximimum throughput, reduced job pending and turnaround times, and faster response times for large or latency-sensitive applications.


Univa – Advanced Services for Intel Omni Path Clusters

With a broad array of products and services, Univa offers the advanced capabilities you need to get the most out of your Intel Omni-Path investment. Univa is equally at home supporting traditional HPC workloads, distributed frameworks and containerized workloads.

Univa Grid Engine – The leading distributed resource management system that manages workloads automatically, maximises shared resources and accelerates run-times Unisight Reporting – Comprehensive monitoring, reporting and analytics to measure and track utilization and performance and remove bottlenecks for optimal performance License Orchestrator – Efficiently manage EDA, CAE and other ISV application licenses on Intel Omni-Path clusters for maximum return on software investments Universal Resource Broker – An open-source solution that extends your Intel Grid Engine or Kubernetes cluster to support additional Big Data frameworks and data center services Container Edition – run Docker Containers seamlessly, blending container and non-container workloads under Grid Engine control Univa Service and Support – Tailored assistance with cluster planning, deployment, application integration and world-class support, delivered by Univa's distributed computing experts

Contact us, and learn how Univa can help you get take full advantage of your Intel Omni-Path investment.