- Leading HPC Team to deliver HPC Compute services for over 800 research users.- Managing multiple HPC (DELL) and GPU (A100s) Clusters with SLURM Job scheduler, IB network and Vast storage for scratch space.- Implementing in-band and out-of-band infrastructure monitoring solutions (Checkmk) to ensure the reliability, performance, and security of EMBL-EBI HPC systems.- Deploying reporting interfaces (XDMoD) for SLURM Job scheduler to get visibility into infrastructure usage and… Show more - Leading HPC Team to deliver HPC Compute services for over 800 research users.- Managing multiple HPC (DELL) and GPU (A100s) Clusters with SLURM Job scheduler, IB network and Vast storage for scratch space.- Implementing in-band and out-of-band infrastructure monitoring solutions (Checkmk) to ensure the reliability, performance, and security of EMBL-EBI HPC systems.- Deploying reporting interfaces (XDMoD) for SLURM Job scheduler to get visibility into infrastructure usage and resource utilisation.- Optimising provisioning methods by introducing stateless provisioning using Warewulf.- Enabling remote visualization to minimize data transfers and enable users to view results in close proximity to the computing and storage resources. Show less

- Facilitating HPC (Storage and Compute) Infrastructure Operations.- Working in close collaboration with various stakeholders to facilitate end users computation and data pipelines.- Budgeting and procuring hardware and software for HPC Center needs.- Planning, Procuring and Deployment of on-perm OpenStack cluster for HPC users to facilitate Data Simulation and ML pipelines.- Managing projects related to HPC infrastructure expansion.- Consulting users on Scientific code… Show more - Facilitating HPC (Storage and Compute) Infrastructure Operations.- Working in close collaboration with various stakeholders to facilitate end users computation and data pipelines.- Budgeting and procuring hardware and software for HPC Center needs.- Planning, Procuring and Deployment of on-perm OpenStack cluster for HPC users to facilitate Data Simulation and ML pipelines.- Managing projects related to HPC infrastructure expansion.- Consulting users on Scientific code development. Show less

- Successfully administrating 300+ Linux based HPC servers for Artificial intelligence and Scientific computing applications. This includes; Cray XC40 HPC system, Nvidia V100 based AI cluster, Hadoop Cluster and Bare metal Linux servers for data scientists. In past, I have experience managing Bull and SGI based HPC systems. I am also managing high throughput storage systems for big data analytics.- Architected and deployed Nvidia V100 based GPU cluster for AI and Data Science… Show more - Successfully administrating 300+ Linux based HPC servers for Artificial intelligence and Scientific computing applications. This includes; Cray XC40 HPC system, Nvidia V100 based AI cluster, Hadoop Cluster and Bare metal Linux servers for data scientists. In past, I have experience managing Bull and SGI based HPC systems. I am also managing high throughput storage systems for big data analytics.- Architected and deployed Nvidia V100 based GPU cluster for AI and Data Science workloads. Integrated system with already existing infrastructure and DDN ExaScaler storage system. By using SLURM workload manager, the system was configured to allow multi tenancy on same host while isolating GPUs per user.- Promoted use of GPU cluster among user community by organizing trainings and identifying use cases. Consulting users on their research projects and assisting them in parallel programming to port their workloads on HPC system.- Developing workflows and tool chains for different Scientific and AI applications. Working closely with developers to integrate scientific codes in optimized workflows so to achieve maximum performance out of each core on the HPC system. This also includes compiling optimized scientific packages for GPU and Compute infrastructure.- Containerizing various Scientific Computing and AI applications and porting them on AI infrastructure. Also, supporting users to build their own data science application inside a container and port it to the system.- Working with Scientific Computing group to consult users with Scientific Computing, Parallel Programming, High Performance Computing and GPU Computing. Show less

- Delivered practical labs on Linux, C/C+, Assembly Language, Shell Scripting, Parallel Programming and Operating Systems. Mentored students on different Linux based projects related to distributed computing.- 80% of labs were Linux based. Worked closely with lab administrators to make sure that the required software stack is available for the students.