Run millions of jobs per day with HPC Works Velocity using scalable, event-driven job scheduling, with sub-millisecond latency for maximum throughput in demanding industries like semiconductor design.
Make the most of expensive electronic design automation (EDA) software and maximize your computing resources, from small, dedicated server farms to complex, distributed HPC, cluster and cloud environments. Boost throughput further with hierarchical scheduling.
Maximize capacity utilization and accelerate throughput with event-driven architecture, a small memory footprint and extremely low dispatch latency. Stay informed with a big-picture view of completed, failed and waiting jobs.
Enable critical jobs to run before less important jobs using features such as preemption and reservations. Suspend running jobs and resume them once higher-priority jobs have completed, so you don’t need to set licenses aside.
Combine fast enterprise job scheduling with industry-leading tools for design flow visualization and dependency management, detailed monitoring and reporting, license allocation and more.
Control workloads, ensure fair user access and prioritize critical jobs by setting and enforcing policies using features such as fairshare, reservations and preemption. Optimize your workload environment to meet changing needs with clear, detailed job visibility that informs data-driven policy decisions.
Optimize workloads with high-throughput, cost-conscious job scheduling in the cloud. Quickly and seamlessly scale up your computing resources. Bring cloud costs closer than ever to exact demand by scaling cloud resources back down to zero when they’re no longer needed.
Increase throughput and improve resource utilization by offloading the base scheduler with hierarchical scheduling. Millisecond dispatch latency benefits short jobs, which users can queue sequentially. Reduce the lower-level scheduling burden by viewing a batch of short jobs as one longer job. Deployments have realized a 6-10x increase in throughput with no added compute capacity.
