change
starts now...

Responsibilities:

• Administration of HPC clusters and parallel file systems
  
• Architecting innovative solutions to complex operating and file system issues
  
• Linux cluster troubleshooting
  
• Technology evaluation and selection
  
• Providing direct support and education of the researcher community
  
• Job scheduler optimization
  
• Deep dives into specific cluster workloads to optimize them
  
• Management and optimization of many petabytes of storage
  
• Build and support non Beowulf clusters (Hadoop, Spark, etc.)
  

Experience & Skills Required:

• Excellent Linux systems administration and troubleshooting skills
  
• Experience in design and engineering of HPC systems and schedulers (Slurm, SGE, TORQUE)
  
• Experience administering and tuning scale out storage (GPFS, BeeGFS, Lustre, Ceph) 
  
• Experience with provisioning systems (Werewolf, Rocks, Bright, xCat, Mesos)
  
• Advanced scripting or programming skills (C/C++, Python, Perl, or UNIX shell)