HPC Consultant

• Responsible for designing, optimizing, and supporting high-performance computing (HPC) environments including cluster scheduling, storage performance, application optimization, and system tuning.
• The role involves improving workload efficiency, supporting HPC applications, and ensuring optimal performance across compute, storage, and network layers in large-scale production environments., * Design, configure, tune, and optimize SLURM partitions, queues, QoS, and scheduling policies.
• Analyze job scheduling behavior, bottlenecks, and resource contention issues.
• Troubleshoot job failures and performance degradation in HPC environments.
• Implement scheduling policies such as fair-share, backfill, and reservations.
• Lead HPC storage benchmarking and performance validation activities.
• Analyze HPC workload I/O patterns and recommend storage architectures.
• Support storage procurement decisions including performance and sizing analysis.
• Collaborate with vendors and internal teams during proof-of-concept evaluations.
• Build, configure, and maintain HPC applications, compilers, and software stacks.
• Optimize application performance using MPI, OpenMP, and GPU acceleration where applicable.
• Manage environment modules and software management frameworks.
• Perform system-level tuning across compute, memory, network, and storage systems.
• Diagnose and resolve node-level issues involving CPU, GPU, interconnects, and OS configurations.
• Create runbooks, performance baselines, and troubleshooting documentation.
• Support cluster upgrades, expansions, and infrastructure lifecycle activities.
• Collaborate with researchers, application owners, and infrastructure teams.
• Translate workload requirements into optimized HPC configurations.
• Provide technical guidance and recommendations to stakeholders and leadership.

• Eight to twelve years of hands-on HPC engineering experience in production environments.
• Strong expertise in SLURM configuration, tuning, and troubleshooting.
• Strong knowledge of Linux operating systems.
• Experience with HPC storage systems and I/O performance analysis.
• Experience building, installing, and optimizing HPC applications and scientific software stacks.
• Experience with MPI, OpenMP, and HPC toolchains.
• Strong scripting skills in Bash and Python.
• Experience with performance analysis and debugging tools.
• Strong understanding of HPC system architecture and workload optimization.

Experience:

• Experience designing and tuning HPC cluster scheduling policies including fair-share, backfill, and reservations.
• Experience in HPC storage benchmarking using tools such as IOR, FIO, MDTest, and IOzone.
• Experience analyzing I/O patterns and mapping workloads to storage architectures.
• Experience supporting application optimization using compilers and libraries.
• Experience in system-level performance tuning across compute, storage, and network layers.
• Experience supporting cluster upgrades, expansions, and hardware refresh activities., * Experience with GPU-based HPC workloads (CUDA, ROCm).
• Exposure to cloud HPC environments (Azure, AWS, Google Cloud Platform).
• Experience with parallel file systems such as Lustre or IBM Spectrum Scale.
• Experience working with vendors for HPC hardware and storage evaluations.

Skills:

• SLURM scheduling and cluster management.
• Linux system administration.
• HPC storage and I/O performance tuning.
• MPI and OpenMP programming models.
• HPC compilers and toolchains (GCC, Intel, NVIDIA HPC SDK).
• Performance analysis tools.
• Python and Bash scripting.
• Environment modules (Lmod).
• HPC system architecture and optimization.
• GPU computing (preferred).

Qualification And Education:

• Bachelor s or Master s degree in Computer Science, Engineering, or related field preferred.