Nov 2020 – Mar 2025 · 4 yrs 4 mos · Chicago, Illinois, United States

• New HPC System designing, development, and Operation
• Installation/Deployment/maintenance/troubleshooting of the new & existing HPC hardware (CPU/GPU) system and software
• Installation/Deployment/maintenance/troubleshooting the GPFS/Spectrum Scale, ZFS,
• Installation/Deployment/maintenance/troubleshooting of the Backup and Archive solution
• Installation/Deployment/maintenance/troubleshooting the Network (Ethernet& InfiniBand)
• Installation/Deployment/maintenance/troubleshooting the Virtualization/ DevOPs/Hyperscale
• Installation/Deployment/maintenance/troubleshooting the HPC security tool like Crowd-Strike, Rapid7 Carbon- black & Wazuh etc
• Installation/Deployment/maintenance/troubleshooting: - Globus, SAMBA, CIFS, NFS, Ganesha
• Installation/Deployment/maintenance/troubleshooting XCAT/Confluent, Slurm, Check-MK, ThinLinc, 2FA Duo Factor authentication, Various Kind of License servers, Git,
• System Firewalld, Gpfs Snapshot policy, etc
• Upgrade the HPC system firmware (Switch, Hardware), OS patch, Security Patch, etc
• Creating scripts for automating daily routine tasks & health checks by using ansible
• Coordinating & following different vendors during Hardware issues, deployment of the new HPC system Hardware, Software, and infrastructure-related issue resolution
• Verify the vendor Benchmarking submitted results on different architectures like INTEL, AMD, (GROMACS, LAMMPS, QE MgO, QE Ice, HipMCL, HPCG, HPCC, IMB)
• Providing day to day technical support to the users’
• Maintaining/Writing/Update all system-related issues and operational documents on Github.
• Manage different vendor’s HPC servers (IBM/DELL/Lenovo/NVIDIA/DDN/SM, GPUs)

Jan 2020 – Oct 2020 · 9 mos · New Delhi, Delhi, India · On-site

• Responsibilities: -
• Manage team responsible for providing on-site hardware, systems, sub-systems, and/or other applications support for customers according to contractual service levels.
• Plan, direct, and monitor operational/tactical activities of the team.
• Meet business and operation targets.
• Recruit and support the development of direct staff members.
• Handle customer escalations.
• Establish relationships with customers and other functional managers.
• Provide guidance on process improvements and recommend changes in alignment with
• Provide coaching and leadership to assigned field technicians.

Oct 2017 – Jan 2020 · 2 yrs 3 mos · Noida Area, India

• CRAY XC40 HPC System: -
• CRAY XC40 2.8 P/F HPC: - CRAY XC40 HPC (Intel Xeon Broadwell E5-2695 V4 18C 2.1
• GHz with 128GB), connected through Cray Aries Dragonfly topology, Total Peak performances 2.8 P/F with 290 TiB memory, home 6.3 PB & 670 TB scratch storage & TFinity SPECTRA Logic Library (19 PB capacity with LTO 7)
• Total 2320 compute nodes (83520 cores)
• Responsibility: -
• Managing & monitoring on-site installation, service & repair of Cray XC40 Cluster machine components Providing 24*7 Maintenance support of 2.8 petaflop HPC machine.
• HPC Site Team lead of 2.8 P/F machine
• Maintaining Site SLA & providing 24*7 support
• Update the system firmware to the latest version (O.S, storage, switches & library, etc)
• A part of the team member during the commissioning & installation of the Indian Biggest HPC Project
• Managing the Datacenter infrastructure
• Lead the localization during UAT & NOA
• Managing user account and quota policy, production jobs & PBS queue as per client requirement.
• Customize the Node image through BCM (Bright Cluster Manager)
• Backup & migration of production data through TAS VSM using spectra Tfinity library
• Involved in day-to-day maintenance & client requirement activity
• Installation/management of tools like Matlab, CDT, IDL, SVN, FlexNet, etc.
• Support weather forecasting models like UM, WRF, NCUM, and GFS
• System configuration and management through Automation tools like Ansible and Puppet
• Documentation of every system administrator procedure
• Coordinate with scientists to change and implement the new models in system
• Follow-up with infrastructure vendors

Mar 2013 – Sep 2017 · 4 yrs 6 mos · Noida Area, India

• IBM IdataPlex 350 T/F HPC Cluster:-
• Bhaskara IBM iDataPlex DX360M4(Xeon E5-2670 8C 2.6 GHz, Infini- Band FDR 14) High-performance computing system 350TF with 67 TB aggregate distributed shared Memory with 3 PB of Storage (India’s most powerful supercomputer in 2015). Total 1052 Computes nodes (16832 cores).
• Responsibility:
• Implementation and maintenance of 350T/F machine.
• Installation and Managing General Parallel File System Cluster (GPFS).
• Installation and configuration of Job Scheduler-IBM LSF.
• Installation and configuration of Platform Application Central for Monitoring and GUI job submission.
• Installation and Up-gradation of Scientific Application & Libraries on RHEL Cluster.
• Porting Regional and Atmospheric Models (UM, NGEFS, WRF) on RHELCluster.
• Creating, Modifying, and distributing User ID across Computing nodes via XCAT.
• Install, maintain and support hardware and software components of the HPC environment.
• Provide immediate support in implementation, troubleshooting, and maintenance of the HPC system.
• Installation and configuration of intel cluster studio, IBM POE & required Modules.
• Manage and troubleshoot TSM & HSM servers and clients.
• Install and troubleshoot Mellanox InfiniBand FDR 14, Cisco MDS9148 Fabric Switches.
• Troubleshooting of TS3500 & Ts1060 Model Tape Drives & library.
• Manage and troubleshoot XCAT, LSF, IDL, and MATLAB Server applications.
• Experiences in installation & system performance checking through LINPAC, BLAS, HPL, Intel MKL Bench-marking Application.

Jan 2012 – Feb 2013 · 1 yr 1 mo · Noida Area, India

• IBM P6-575 22T/F HPC System: -
• Environment: IBM P6-575 22TF HPC (Each Node with 32 cores of IBM P6 CPU 4.7 GHz with 32GB RAM), a Total of 42 Nodes connected through the InfiniBand network. Peak Speed ~ 22 TFs with ~500TB GPFS storage & IBM 3584 Tape library of 100TB capacity (LTO4)
• Responsibility: -
• System hardware monitoring, support & repair of IBM P6 (P575) based HPC components and other IBM P-series servers (p560, p570) using Hardware Management Console (HMC).
• Log’s collection, identification, Disk Management & replacements in IBM Storage DS4k.
• IB cable errors, issue tracing, replacements & troubleshooting in InfiniBand Q-logic Switches.
• Production Job Monitoring, Scheduling & configuration using IBM workload scheduler: LoadLeveler.
• AIX v5 & v6 Operating System related issues identification & troubleshoot.
• Support Weather Forecasting models like UM, WRF, NCUM & GFS.
• Data Backup & Migration using IBM TSM (Tivoli Storage Manager) in LTO4 Tape drives.
• Troubleshooting Network & Booting problems, Monitoring & performance optimization.
• User creation & Modification with file system quota.
• Generating Daily, weekly & monthly reports of System utilization using the NMON tool.