Univa

Univa
Nbendov · CC BY-SA 4.0 · source
Name	Univa
Type	Private
Industry	Software
Founded	2004
Fate	Acquired by [See Company Structure and Corporate Affairs]
Headquarters	Ann Arbor, Michigan
Products	Grid Engine, Univa Grid Engine, Navops Launch

Univa was a software company specializing in workload management and cluster resource scheduling for high-performance computing environments. Founded in 2004, the firm developed commercial distributions and support around grid and cluster management technologies that served research institutions, corporations, and national laboratories. Univa's offerings integrated with a range of hardware vendors and software ecosystems to manage compute, storage, and cloud resources.

History

Univa was founded by former employees of a project tied to Sun Microsystems, Oracle Corporation, and the open-source Grid Engine community. The company emerged amid transitions involving Sun Grid Engine, the acquisition of Sun Microsystems by Oracle Corporation, and controversies over stewardship of the Grid Engine codebase. Early milestones included commercial releases aimed at customers migrating from Sun Grid Engine and organizations such as Lawrence Livermore National Laboratory, Argonne National Laboratory, and many university research computing centers. In the 2010s Univa announced programs to support migrations from Platform LSF and IBM workloads, while partnering with hardware vendors like Dell Technologies, Hewlett Packard Enterprise, and Intel Corporation. The company later expanded into cloud orchestration with integrations for Amazon Web Services, Microsoft Azure, and Google Cloud Platform. In a strategic corporate transaction, assets and personnel were integrated into another firm in the late 2010s, marking the end of Univa as an independent operator.

Products and Technology

Univa offered commercial versions of workload schedulers compatible with legacy Grid Engine deployments and modern high-performance computing stacks. Core commercial products included Univa's distribution of the Grid Engine scheduler and a cloud-focused orchestration product that automated provisioning on Amazon EC2, Azure Virtual Machines, and Google Compute Engine. The company provided integrations for middleware and platforms such as Hadoop, Spark, Slurm Workload Manager, Kubernetes, and MPI implementations like Open MPI and Intel MPI. Univa also offered monitoring and reporting integrations with tools and projects including Nagios, Ganglia, and Prometheus to provide telemetry and accounting for customers like National Center for Supercomputing Applications and European Organization for Nuclear Research users.

Architecture and Features

Univa's architecture centered on a master/agent scheduler model derived from the Grid Engine lineage, supporting features such as job queuing, resource reservations, fair-share policies, and advanced scheduling algorithms. The product suite supported heterogeneous hardware from vendors like NVIDIA, AMD, and IBM Power Systems and accommodated accelerators such as NVIDIA Tesla and FPGA cards from Xilinx. Key features included multi-cluster federation for geographically distributed deployments (used by consortia such as Open Science Grid), checkpoint/restart support via integrations with libraries like BLCR, and license-aware scheduling compatible with vendors such as Synopsys and ANSYS. For cloud deployments, capabilities included spot and preemptible instance handling consistent with AWS Spot Instances and Google Preemptible VM models, autoscaling policies, and billing integrations with platforms like OpenStack and VMware vCloud.

Deployments and Use Cases

Univa's customers spanned academic research, government laboratories, and commercial enterprises. Typical deployments were found at institutions such as Lawrence Berkeley National Laboratory, Sandia National Laboratories, Los Alamos National Laboratory, and major universities participating in research projects funded by agencies like the National Science Foundation and the Department of Energy. Use cases ranged from computational fluid dynamics on clusters from Cray and HPE Cray to genomics pipelines leveraging tools tied to Broad Institute workflows and pharmaceutical computational chemistry used by firms like Pfizer and GlaxoSmithKline. Financial services firms using risk modeling on clusters from IBM relied on Univa scheduling to integrate with analytics stacks including MATLAB, R Project, and SAS. In media and entertainment, render farms employing software such as Autodesk Maya and Pixar RenderMan used Univa for job distribution and license management.

Company Structure and Corporate Affairs

Univa operated as a privately held company headquartered in Ann Arbor, Michigan, with sales and engineering teams distributed across North America and Europe, and partnerships with system integrators like Wipro and HCLTech. The company's leadership included executives with backgrounds at Sun Microsystems, Oracle Corporation, and other enterprise software firms. Univa engaged in commercial support contracts, professional services, and OEM agreements with vendors including Dell EMC and Hewlett Packard Enterprise. In a notable corporate development, Univa's product line and personnel were acquired and assimilated into another organization, aligning with consolidation trends in the enterprise workload management market alongside firms such as Altair Engineering, IBM, and SUSE. The acquisition influenced customers' support pathways and product roadmaps across the high-performance computing ecosystem.

Category:Software companies