Univa®, a leading innovator in enterprise-grade workload management and optimization solutions, today announced enhancements to its powerful Navops Launch HPC cloud-automation platform to support the widely-used Slurm workload scheduler.
Slurm (Simple Linux Utility for Resource Management) is a popular open-source workload manager supported by SchedMD that is well known for its pluggable HPC scheduling features. Slurm is used by approximately 60% of the world’s Top 500 supercomputers and is used by 45% of HPC cloud deployments.
“Univa’s support for open-source Slurm – installed at approximately 60% of the leading HPC centers – is an important milestone in the evolution of HPC cloud-automation,” said Srini Chari, founder and managing partner of Cabot Partners. “Navops Launch is one of the few HPC management tools that is both multi-cloud and multi-scheduler.”
By using Navops Launch with Slurm, users can easily migrate workloads to the cloud, control spending, rightsize cloud resource selection and automate decisions based on real-time cloud, application, budget and workload-related metrics using composable automation applets.
“Extending Navops Launch support to Slurm is a natural fit given our ability to run large-scale HPC workloads,” says Gary Tyreman, Univa’s President and CEO. “Navops Launch can help Slurm users simplify cloud and hybrid cloud deployments and dramatically reduce costs by ensuring that HPC sites only pay for the resources they need.”
Enhanced Cloud Capabilities for Slurm Users
While Slurm presently supports simple cloud bursting via an elastic power plugin and cloud-specific integrations, Navops Launch brings rich new functionality including flexible policy-based automation, multi-cloud support, data migration, and cost control with cloud spend association. Unlike tools that track cloud resource usage and application workloads separately, Navops Launch monitors both and has full visibility into how Slurm is using cloud resources in real-time. This allows Navops Launch to take policy-driven actions in real-time such as shifting workloads to the cloud, scaling services, moving data to lower-cost storage tiers, or taking advantage of spot, preemptible or best-priced instances to boost efficiency and ensure that cloud spending targets are met.
Built for Large-Scale HPC
In early 2019, Western Digital, supported by Univa and Amazon Web Services, used Univa Navops Launch to deploy a million-core cluster in just 1 hour and 32 minutes. The deployed cluster ran a complex multiphysics simulation that normally required 20 days to complete in just 8 hours – a 60x performance improvement.
Product and Support Availability
Slurm support will be available in Navops Launch in the Fall of 2019 as part of Navops Launch 2.0. At that time, Slurm users will have access to the same cloud automation capabilities available to Univa Grid Engine users today.
About Navops Launch
Navops Launch is designed to avoid bottlenecks such as cloud API calls, resource monitoring, and name service scalability that typically plague large-scale cloud-based cluster deployments. Navops Launch is production-proven at many of Univa’s enterprise hybrid cloud customers. Additionally, Navops Launch offers users the ability to deploy and manage hybrid and dedicated HPC clusters for thousands of applications, including AI, machine learning, life sciences, manufacturing, and transportation.
About Univa Corporation
Univa is the leading innovator of workload management solutions that optimize throughput and performance of applications, containers, and services. Univa manages workloads automatically by maximizing shared resources and enabling enterprises to scale compute resources across on-premise, hybrid and cloud infrastructures. Univa’s solutions help hundreds of companies to manage thousands of applications and run billions of tasks every day to obtain actionable insights and achieve faster time-to-results. Univa is headquartered in Chicago, with offices in Canada and Germany.