New solution reduces HPC deployment time by 90% while maintaining native Slurm functionality and Kubernetes flexibility and full operational visibility
SAN FRANCISCO, CA / ACCESS Newswire / November 18, 2025 / ClearML, the leading end-to-end solution for GPU management and unleashing AI in the enterprise, today launched one-click dynamic Slurm cluster deployment on Kubernetes infrastructure. The new solution eliminates the operational overhead and inefficiencies maintaining dedicated high-performance computing (HPC) clusters with Kubernetes requirements for AI workloads. With ClearML Slurm deployment, IT admins can deploy dynamically scalable Slurm clusters for HPC workloads with scale-to-zero autoscaling and monitoring capabilities absent from mixed Slurm and Kubernetes environments.
By enabling Slurm scheduling on Kubernetes infrastructure, ClearML allows organizations to effortlessly maintain existing workflows while gaining the benefits and flexibility of secure Kubernetes orchestration. Enterprises benefit from elastic clusters with dynamic resizing, including scale-to-zero when not in use, while GPU passthrough orchestration enables workloads to achieve bare-metal performance, avoiding the need to manage a dedicated Slurm cluster.
This new solution, available within ClearML’s Infrastructure Control Plane, provides a dashboard with real-time visibility into job status, execution history, and failure diagnostics while enabling secure multi-tenant deployments with dynamic Slurm clusters and fully secure separation of resources, network and storage for each tenant. This expanded capability for workload management reduces the overhead for IT teams while also providing flexibility for AI builders and traditional HPC workload management.
“HPC teams shouldn’t have to choose between the Slurm workflows they trust and the flexibility of cloud-native infrastructure,” said Moses Guttmann, CEO & Co-founder of ClearML. “We’re eliminating the administrative burden of dedicated Slurm clusters while adding the autoscaling, monitoring, and multi-tenancy capabilities that HPC teams need as their computational demands evolve.”
With ClearML, HPC and AI workloads can operate on shared infrastructure unobstructed, addressing the growing need of enterprises and organizations running both traditional HPC workloads such as physics simulations in conjunction with AI workloads.
ClearML also launched one-click deployment of multi-node AI training, simplifying distributed training operations across any hardware infrastructure. This new capability automatically orchestrates codebases based on frameworks such as pytorch distributed and provides a single dashboard into parallelized training workloads including visibility into training progress, the number of running nodes, and their statuses. Organizations can leverage the feature for fine-tuning large language models, training foundation models, and accelerating training cycles through parallel resource allocation. With this new solution, administrators can dynamically configure the resources available to the distributed training workloads which include native autoscalers that spin down idle instances after a set time. ClearML’s one-click multi-node training deployment makes it easy to launch distributed training to accelerate model training and fine-tuning.
Experience Live Demos at SC25
Both new solutions will be demonstrated live at the SC25 Conference (Supercomputing 2025) in St. Louis, Missouri, November 17-20. Visit ClearML at booth #3209.
About ClearML
As the leading infrastructure platform for unleashing AI in organizations worldwide, ClearML is used by more than 2,100 customers to manage GPU clusters and optimize utilization, streamline AI/ML workflows, and deploy GenAI models effortlessly. ClearML is an NVIDIA partner and is trusted by more than 300,000 forward-thinking AI builders and IT teams at leading Fortune 500 companies, enterprises, academia, public sector agencies, and innovative start-ups worldwide. To learn more, visit the company’s website at https://clear.ml.
Media Contact
Noam Harel
Chief Marketing Officer
ClearML
[email protected]
SOURCE: ClearML, Inc.
