Gridware Cluster Scheduler
Established in 2024, HPC Gridware’s mission is to transform HPC and AI workload management into a scalable, user-friendly, and affordable experience tailored to meet the demands of organizations of any size—including enterprises in sectors like pharmaceuticals, electronic design automation, automotive, and finance. HPC Gridware’s platform delivers robust flexibility and custom features to precisely match your cluster requirements, ensuring efficient utilization and reliability for any workload, from traditional HPC to cutting-edge AI.
Gridware Cluster Scheduler is a powerful workload manager developed by HPC Gridware GmbH, built on the Open Cluster Scheduler —an open-source workload manager formerly known as "Sun Grid Engine". Designed specifically for high-performance computing (HPC) and AI, it ensures your on-premises or cloud clusters remain highly utilized around the clock.
Thanks to full SGE compatibility, the scheduler seamlessly supports all application types, from single-process and single-node jobs to complex multi-node MPI workloads, across a wide range of architectures such as AMD64, ARM64, and RISC-V, with full GPU support and compatibility with popular MPI implementations. Advanced automation features and feature-rich user interface make it simple to manage, monitor, and scale complex workloads, empowering both administrators and end users with efficient and reliable operations.
Why Gridware Cluster Scheduler is
SimOps compliant
In the following, we check the SimOps compliance of the Gridware Cluster Scheduler.
With these capabilities, Gridware Cluster Scheduler provides a robust, future-proof foundation for any organization implementing SimOps, empowering teams to maximize the value, efficiency, and reliability of their simulation-driven operations.
1. Universal simulation support
Gridware Cluster Scheduler offers full "Grid Engine" compatibility, enabling it to run any simulation tool—including all MPI implementations and supporting CPUs, GPUs, and multiple architectures—so organizations can adopt a simulation-driven approach to product development.
2. Workflow automation and resource optimization
Advanced features such as resource reservation, job arrays, inter-job dependencies, dynamic scheduling, and intelligent workload distribution make it easy to automate, scale, and optimize complex simulation workflows, ensuring maximum efficiency and minimal idle time.
3. Cloud operational flexibility
Gridware Cluster Scheduler can manage workloads seamlessly on-premises, or in the cloud, delivering the flexible and consistent operations required by SimOps best practices.
4. Observability and actionable insights
Integration with cloud-native monitoring tools like Prometheus and Grafana ensures real-time visibility into resource usage, job states, and cluster health, empowering data-driven decision-making and fast troubleshooting.
5. Automated management and cost control
Powerful and extensible job accounting, and automated job management allow for precise tracking and optimization of costs, power usage and resources, fully supporting the SimOps objective of balancing resource utilization with operational efficiency.
