Linux System Administrator
CurrentStaff with TechSquare, Inc. on-site at the MGHPCC in Holyoke, MA.Duties currently include projects and administration for MIT's ORCD (Office of Research Computing and Data): - HPC system administration - Managing server and networking hardware installations - Extensive server and network cabling throughout the datacenter - Provisioning servers and switches for use within ORCD clusters - Working with vendors such as Nvidia, Dell, EchoStor, Microway, Thinkmate on projects and build-outs - Inventory tracking and reconciliation - Troubleshooting server issues - Responsible for coordinating the recycling of old compute and network hardware - Care of MIT's Satori, Engaging and Openmind clusters, such as: -- Checking daily logs and the state of compute nodes via Slurm -- Ensuring health of backups and filesystems -- Reviewing NOC for issues -- Resolving requests regarding quota/QoS changes for cluster users -- Cluster software and job submission troubleshooting -- Installing software, creating environment modules Previous responsibilities have included:- Working with Northeastern's Research Computing admins to: -- Provide datacenter administration remote-hands -- Implement updated Infiniband fabric configuration -- Expand current working clusters -- Deprecate old hardware- Hardware remote hands for the UMass Unity/Gypsum cluster: -- Including server troubleshooting for the University of Rhode Island- Ad-hoc work for other clients such as Harvard, BU, NESE, Boston Children's Hospital, Psivant (Roivant), and numerous labs and groups at MIT such as CSAIL- Script writing (Bash, Python)- Creating and testing Open OnDemand interactive apps- Thorough documentation of procedures related to all work- Testing package installation with SaltStack on different Linux distributions- Aiding in the migration of a locally-used web app to an internal company server