HPC Administrator
United for Manpower Solutions
- Doha, Qatar
- Contract
- Full-time
- Installs software on HPC systems writes test scripts, troubleshoots application problems, and runs benchmarks to evaluate the performance of algorithms on different configurations.
- Assists in managing and monitoring the performance and stability of the HPC resources.
- Oversees the health, compliance, and performance of various HPC systems.
- Contributes to the design and configuration of HPC systems in response to business requirements.
- Experience in building and managing containers using different technologies like Docker and Singularity.
- Experience setting up and maintaining scientific computing clusters and their associated scheduling systems, o such as LSF, or Slurm
- Experience setting up and maintaining a clustered file system such as GPFS or others.
- Provides monthly reports on the performance and utilization of the HPC systems to their management.
- Installs software & manages file systems and troubleshoots alerts from monitoring tools.
- Assist researchers with debugging problems that arise when compiling- using HPC resources or linking to HPC-specific libraries (for example, C, C++, Matlab, Perl, R, openmp, mpi, cuda, pthreads, etc.