Souvik Paul

CEO

Pune, Maharashtra, India12 yrs 4 mos experience

Most Likely To SwitchHighly Stable

Key Highlights

Over 8 years of experience in HPC administration.
Expertise in deploying and managing large-scale HPC systems.
Strong knowledge of Linux environments and scripting.

Stackforce AI infers this person is a Supercomputing Infrastructure Specialist with extensive HPC management experience.

Contact

Skills

Core Skills

Hpc AdministrationCluster Management

Other Skills

Backup Tape StorageCentOSCorosyncDRBDIBM Platform HPCIDSInfinibandIntel ArchitectureIntel KNLCoprocessorsLSF SchedulerLustre StorageNetwork SecurityNmapNvidia P100OpenHPC

About

Work experience of around 8 years 6 months in High performances computing (HPC) administration and Cluster Management based on Linux • Thorough understanding of HPC Environments. • Integration and deployment of HPC Systems. • Good knowledge on Linux Environments on Intel x86/64 and Arm based processors. • Good knowledge of filesystems like Lustre/LVM. • Strong knowledge on TCP/IP • Comprehensive Knowledge with gdb, analyzing stack trace. • Automate tasks in HPC environment in bash and Python programming language. • Good knowledge of Shell and Python scripting. • HPC workload managers like PBS Professional and Slurm. • Good Knowledge on Lustre HSM and Tape Based Storage. • Co-processors: NVIDIA GPU (A100, P100, K40, Tesla K20) , Intel MIC • InfiniBand and Omni path Interconnects - Configuration and troubleshooting. • Directory service: Open LDAP, NIS • Benchmarks: HPL, IOZONE, IOR, Iperf • Good knowledge on configuration management with Ansible. • Hardware: Cray, HPE Apollo, Dell, Fujitsu, Super Micro • Storages: • DDN Lustre, Cray ClusterStor, TAS with Spectra Tfinity, Cray Sonexion,Netapp ,VSM, SAN and NAS based storage

Experience

Applied materials

HPC Team Lead

Apr 2022 – Present · 3 yrs 11 mos · India

Hewlett packard enterprise

System Specialist

Jan 2020 – Apr 2022 · 2 yrs 3 mos · Pune, Maharashtra, India

Cray inc.

System Engineer

Nov 2017 – Dec 2019 · 2 yrs 1 mo · Pune Area, India

Deployment and Maintenance of XC40 (Pratyush) comprising of 3327 nodes of 4PF having 10 PB of Lustre Storage and 27.4 PB of Backup Tape based Storage and XC50 Clusters comprising of 44 nodes at IITM Pune.

Cdac r&d

Project Engineer

Sep 2013 – Oct 2017 · 4 yrs 1 mo · Pune Area, India

Projects:
HPC Cluster Deployment Projects
Project: Deployment of Param Servers at HPC Technologies
Software: OpenHPC, Slurm Scheduler
Key Project Achievements:
Deployment and maintenance of High Performance Cluster having 40 nodes with Nvidia P100 and Intel KNLCoprocessors.
Worked on Intel Omni-path and Mellanox Infiniband Interconnects having lustre parallel filesytem.
Project: High Performance Computingonsite System Admin at NIT Silchar Client side
Software: IBM Platform HPC,LSF Scheduler with Intel Cluster Studio
Key Project Achievements:
Deployment and maintenance of a High Performance Cluster of 17 nodes with Nvidia Kepler K20x and Intel Xeon Phi Coprocessors.
Took lead on several projects including Vsphere Esxi Virtualization, automation of scripts for daily logs, system monitoring, configuration of iptables, wireshark, tcpdumpfor proper Network Security Measures, Resource and user quota, applying routine OS patches.
Configured IBM Platform HPC, worked on LSF scheduler, Mellanox Infiniband Interconnects with SAN Storage.
Licence management, make changes in make file and configure and install applications
Conducted many training Programmes on Linux Cluster administration from CDAC
Project: Deployment of Param Shavak at Assam Engineering College.
Software: CentOS, PBSJob Scheduler
Build/Rebuild OS, Integration of coprocessor components.Trainings and Workshops on administration and usability of the HPCCluster.
Remote Technical support.
Project: - “High Availability Cluster Computing”
Tools Used: Pacemaker, corosync, apache, drbd, gfs2.
Implemented High Availability Clustering based on Linux environment using open source tools which typically implements concepts like heartbeat monitoring, virtual ip, high availability, clustered services, resource monitoring, drbd,gfs2 and resource monitoring.