Today, we are pleased to announce the general availability of Navops Launch 2.1, with support for Slurm-based clusters. Navops Launch is an automation and spend management platform for migrating compute-intensive HPC workloads to the cloud.
With this release, Navops Launch provides support for both Univa Grid Engine and Slurm-based clusters. These tools manage approximately 54% of HPC cloud workloads, according to a recent InsideHPC survey sponsored by Univa. Slurm is used in leading supercomputers and research organizations and is known for its community support, parallel job capabilities, and an extensive library of plug-ins.
While most Slurm users are no strangers to the cloud, Navops Launch brings a rich array of new cloud migration, automation, and spend management capabilities. Rather than relying on bursting scripts, and cloud-specific integrations, Navops Launch provides a generalized, scalable cloud management foundation with a sophisticated rules-based automation engine.
With Navops Launch, administrators can easily create automations based on a wide variety of metrics related to applications, workloads, and cloud resources. Actions may include scaling cloud resources, migrating workloads, and moving or synchronizing datasets between local and remote filesystems and object stores. Navops Launch helps automate some of the most challenging aspects of operating in the cloud. These innovations combine to help administrators operate more efficiently and obtain higher levels of throughput and productivity while minimizing cloud spending.
An enhanced Web UI provides complete visibility to costs, workloads, and resource consumption efficiency across cost-centers, projects, and departments. In the same survey mentioned above, 84% of HPC cloud users saw value in automating cloud-spend association, but most had no automated solution.[i] The spend management functionality in Navops Launch directly addresses this critical gap.
This release of Navops Launch also provides new functionality for Univa Grid Engine users, including enhanced support for image-based provisioning and support for cloud Scale Sets. With these enhancements, cloud-resident clusters can scale larger, faster, and more cost-effectively. Univa Grid Engine users can also take advantage of new managed tags functionality. This enhancement makes tags assigned to cloud resources in Navops Launch instantly available as requestable resources within Univa Grid Engine, dramatically improving flexibility.
The use of cloud for HPC is dramatically on the rise. Hyperion Research reported in December 2019 that HPC-related cloud spending grew to USD 4 Billion in 2019, a 60% jump over 2018. Moreover, cloud spending is expected to continue growing at a steady ~25% CAGR through 2024.[ii]
We are excited that with Navops Launch 2.1, Univa will have the opportunity to reach a broad new community of HPC cloud users. Navops Launch can be used to deploy and manage hybrid and dedicated HPC clusters for thousands of applications, including AI, machine learning, life sciences, manufacturing, and transportation.
Slurm users can obtain additional information or request a personalized demonstration of Navops Launch 2.1 by contacting us at Univa.com.