HPC / Slurm Infrastructure Engineer

HPC / Slurm Infrastructure Engineer

Contract | Remote | Part-time or Full-time | Rate DOE

I'm working with a technology company looking to improve and scale its engineering compute / HPC environment , with a particular focus on designing and deploying Slurm .

This isn't a requirement for someone who has simply administered an existing Slurm environment or submitted workloads to a cluster. They need someone who has been involved in the architecture, build and deployment of Slurm-based compute infrastructure and can take ownership of the environment from an early stage.

The requirement is still being shaped, so we're looking for someone senior enough to assess the existing infrastructure, recommend the appropriate architecture and then help lead the implementation.

Likely areas of responsibility include:

• Architecture and deployment of Slurm-based HPC / compute clusters
• Slurm controller and compute-node architecture
• Scheduling, queues/partitions and resource management
• QoS, Fairshare and workload prioritisation
• Slurm accounting and monitoring
• Linux compute infrastructure
• Cluster provisioning and configuration
• Infrastructure automation and scripting
• Performance, utilisation and capacity optimisation
• Working with engineering teams to understand compute requirements and translate these into infrastructure

Experience within semiconductor / ASIC engineering environments would be particularly interesting.

Ideally, you'll have supported compute-intensive EDA workloads , such as:

• RTL simulation and verification
• Regression workloads
• Synthesis
• Physical design / implementation
• Cadence, Synopsys or Siemens/Mentor toolchains

The key requirement, however, is genuine Slurm architecture and deployment experience rather than semiconductor experience alone.

If you've previously designed and built Slurm clusters rather than simply used or supported them, I'd be interested in speaking.

Please message me directly for more information.