HPC Engineer/Architect

Job Title: HPC Engineer/Architect Work Location: New York, NY 10001 Job Type: Contract Work Type: Hybrid Duration: 6+ Months USC/GC/H4/H1

Job Summary:

• You will support day-to-day operations of large-scale parallel file systems, deploy and maintain Linux HPC infrastructure across multiple data centers, and assist HPC engineers and architects with day-to-day operations and tickets.
• Support day-to-day operations of large-scale parallel file systems
• Deploy and Maintain Linux HPC infrastructure across multiple datacenters
• Assist HPC engineers and architects with day-to-day operations and tickets

Experience:

• 16 to 20 years

Required Skills:

• Linux Operating Systems (RHEL/CentOS), Parallel file system (GPFS), Job Scheduler LSF/Slrm
• Anxible, Python, Shell scripting
• GPU-based compute infrastructure (including CUDA)
• CentOS 4.5
• HPCC

Responsibilities:

• Design, architect and oversee implementation of Linux based HPC clusters and storage
• Deploy physical hardware using HPC deployment tools and configuration and orchestration tools (Ansible)
• Parallel file system (GPFS) performance tuning, monitoring and troubleshooting
• Perform systems benchmarking, and developing automated tests for the HPC environment, ensuring the reliability and efficiency of our computational infrastructure
• Infiniband network maintenance and troubleshooting
• Automate and monitor the HPC user lifecycle process
• Slurm installation, configuration, performance tuning and troubleshooting
• Plan, design and implement a transition from the LSF scheduler to Slurm
• Manage the Slurm scheduler and translate Research policies into scheduler configurations
• Consult with faculty and students to develop research pipelines for use on the HPC cluster
• Develop and maintain user lifecycle software suite in Python, implement CI/CD pipeline
• Test and automate upgrades of critical system applications using Ansible and shell scripts.
• The ability to communicate effectively with clinicians, researchers, and other team members to develop technological solutions is key

Qualifications:

• Experience working in a large-scale research based HPC environment
• Proven experience working with distributed file storage solutions (i.e., GPFS)
• Experience with deploying and troubleshooting Linux Operating Systems (RHEL/CentOS)
• Experience with Scripting and Automation (Ansible, Python, Shell Scripting)
• Solid understanding of job schedulers (LSF/SLURM)
• Experience with GPU-based compute infrastructure (including CUDA)