Nov 2018 – Dec 2019 · 1 yr 1 mo · Denmark

Nov 2015 – Oct 2018 · 2 yrs 11 mos · Singapore

Jul 2015 – Oct 2015 · 3 mos · Bengaluru, Karnataka, India

Sep 2012 – Jul 2015 · 2 yrs 10 mos · Bengaluru, Karnataka, India

• Currently working with US based Research client for Managing up the Linux HPC
• Cluster Cloud capacity of more than 2000 blade servers and its integration with PANASAS
• NAS storage of 1500+ TB capacity amd providing day to day Operational Support to 1500+
• Researchers in Aviation, Energy, Oil and Gas, Power and Water, Transportation domain.
• Leading a Infrastructure team of 10 members.
• Manage transition from HPC IT Infrastructure Solution implementation to HPC Operational Support.
• Troubleshooting service problems, escalating issues within the organization as needed.
• First level of escalation for production Infrastructure issues.
• Manage production Infrastructure - optimizing availability and high operational “up time”.
• The Configuration, Enhancement, Testing, Integration, Implementation and Documentation of hardware, Operating Systems, Software’s utilizing defined processes and procedures.
• Set technical direction for the team in regards to Infrastructure upgrades and enhancements.
• Project progress status and Operational performance status reporting to HPC Global Operations Leader and HPC Global Project Leader during the Weekly/Pillar meeting.
• Performing hardware refresh in conjunction with OS, software upgrades, etc.
• Providing support for compute and storage hardware, operating systems and utilities.
• Administration of Infiniband (IB) interconnects (DDR, QDR, FDR).
• Managing projects to deploy infrastructure to support HPC Customers.
• Monitoring of cloud HPC Capacity by using Nagios.
• Co-ordination with Vendor for Hardware break/fix for HPC infrastructure.
• Maintain the compute environment with vendor patches, firmware and OS updates.
• Storage management – Coordinate with respective vendors to resolve Panasas storage related issues, manage mount points and access on the cluster.
• PANASAS Storage performance tuning, benchmarking and monitoring the filesystem.
• Adhere to Standard Operating Procedures and defined services delivery processes as provided by client.

Aug 2011 – Aug 2012 · 1 yr · Pune/Pimpri-Chinchwad Area

• 1. EKA 172.6 TF HPC cloud storage/system administration (The 4th fastest Supercomputer in the World according to the TOP500 Ratings released on November 2007)
• Administration and management of EKA (1800+ node Linux cluster)
• Customer lifecycle management – incident, change management
• User requirement gathering
• Proof of Concept setup
• Daily ticket resolution for HPC users
• Storage and backup management
• IB Fabric management
• Scheduler management and optimization – LSF, SLURM, Torque, Maui
• User administration and cluster provisioning
• Parallel file system management – HP Ibrix, Panasas, Lustre, Gluster
• Application commissioning and management
• Preventive maintenance for better uptime
• Adherence to ISO 27001 and ITIL processes
• 2. Upgrade of Luster filesystem from 1.6.7 to 1.8.3 Production storage in CRL
• Setup details:
• Lustre PFS consisting of 10 OSS and MDS/Admin in HA mode
• 38 OST’s with 2TB LUN size
• This PFS serves 1800+ clients simultaneously over Infiniband fabric
• Replicated the original configuration and Test upgrade
• Analysis and resolution of problems
• Data integrity checks and monitoring in early stages
• Client disconnection issue was eliminated post upgrade

Mar 2008 – Aug 2011 · 3 yrs 5 mos · Pune/Pimpri-Chinchwad Area

• 1. PARAM Yuva 54 TF HPC cloud storage administration (The 69th fastest Supercomputer in the World according to the TOP500 Ratings released on November 2008)
• I had the opportunity to be associated with this project right from commissioning of this system till its sustenance mode.
• 2. SAN Cloud Storage Configuration and Administration
• Implementation of HP Storage Area Network, which includes configuring and managing of SAN switches
• Management and administration of HP MSA RAID Arrays
• Zoning, LUN creation, etc.
• Performing firmware upgrades
• Troubleshooting the storage related performance issues
• 3. Parallel filesystem and Backup solution integration for PARAM Yuva
• Installation, configuration and administration of HP Network Storage Gateway server HP X9000 series
• 200+ TB IBRIX appliance was setup and provisioned home area for users
• 200+ TB Lustre was setup for scratch area
• Implemented HP Data Protector backup solution and its integration with HP ESL 712 (670 slots/ 16 drives) Tape Library
• Also responsible for the storage administration of individual storage systems of 100+ TB capacity in the organization.
• Training from HP
• Received configuration and administration training on HP Storage Array, Ibrix filesystem, HP Data Protector backup software, Brocade SAN switches, Lustre, HP ESL Tape Library, Infiniband, etc.