WHPC Poster Authors Awarded Fellowships to Present at HPCKP25 in Barcelona
Safae Bourhnane and Sowmya Shree Awarded HPCKP25 Fellowships to Present at Annual HPC Meeting in Barcelona
Safae Bourhnane, from the Fraunhofer Institute for Algorithms and Scientific Computing (SCAI) in Germany, and Sowmya Shree, from the Centre for Development of Advanced Computing (C-DAC) in India, have been selected to receive presentation fellowships for the upcoming HPC Knowledge Portal (HPCKP25) Annual Meeting, taking place 3–5 June 2025 in Barcelona, Spain.
Bourhnane has been awarded a Travel Fellowship to attend the conference in person, while Shree has received an Online Fellowship Opportunity to participate and present virtually. Both recipients are authors of successfully submitted posters to Women in HPC (WHPC) at ISC25, and were nominated through a highly competitive selection process conducted in collaboration with WHPC. As part of their successful fellowships, both women will present extended talks on their research.
The final selection was made by the HPCKP volunteer committee, who recognized their outstanding potential and contributions to high-performance computing (HPC). The initiative reflects HPCKP’s commitment to supporting early-career researchers and fostering diversity in the HPC community.
The HPCKP Annual Meetingis a key global event that brings together developers, researchers, and professionals in HPC, AI, and Quantum Science. The event serves as a platform for sharing innovations, forming collaborations, and highlighting emerging talent in the field.
The full event agenda is available here, and registration for both in-person and online participation is currently open via this link.
Safae Bourhnane
Safae Bourhnane is a High-Performance Computing researcher at the Fraunhofer Institute for Algorithms and Scientific Computing (SCAI) in Sankt Augustin, Germany. In this role, she develops and evaluates energy-efficient HPC frameworks, with a focus on code optimization and hardware-aware acceleration methods to improve performance and reduce resource consumption.
Safae holds a PhD in Computer Science; her doctoral research investigated energy-efficient HPC alternatives based on Raspberry Pi clusters, demonstrating how low-power, commodity hardware can be orchestrated to support scientific computing workloads.
Prior to joining Fraunhofer SCAI, Safae was an HPC specialist at the Leibniz Institute for Natural Product Research and Infection Biology (Leibniz- HKI). There, she designed and deployed a medium-sized HPC cluster and provided expert support to bioinformatics research groups, enabling scalable analysis pipelines on both on-premises and supercomputing systems.
As founder of the MoroccoHPC initiative, she has worked to cultivate and connect Morocco’s emerging HPC community. Since its launch, MoroccoHPC has organized a number of interactive webinars and workshops, drawing participants from universities, research institutes, and industry across North Africa and Europe.
Topics have ranged from parallel programming and GPU acceleration to workflow automation and cyber-security for HPC systems. The initiative also hosts hackathons and mentorship sessions, providing hands-on training and encouraging collaborations between junior researchers and seasoned practitioners. Through partnerships with academic institutions and national research centers, MoroccoHPC has helped establish a centralized forum for sharing best practices, open-source tools, and case studies—advancing local capacity for large-scale computational science.
Safae will present on metadata management in HPC at the upcoming High Performance Computing Knowledge Portal conference, outlining strategies for capturing and leveraging provenance information to support reproducibility and performance analysis. She will also present a poster and deliver a lightning talk at the International Supercomputing Conference entitled “Streamlining HPC Data Workflows: Cloud SQL Databases at Scale for Metadata.” In this work, she benchmarks PostgreSQL in a cloud environment, evaluating I/O operations, latency, concurrency, and scalability, to assess the suitability of SQL-based databases for managing metadata in large-scale HPC workflows.
Sowmya Shree N
Sowmya Shree N is an HPC/AI Engineer with nearly five years of industry experience, including over three years dedicated to HPC/AI related Projects.
At the Centre for Development of Advanced Computing (C-DAC), she contributes to the National Supercomputing Mission (NSM), where she plays a pivotal role in the deployment, optimization, and validation of HPC/AI workloads on GPU-accelerated systems. Sowmya’s expertise spans benchmarking and fine-tuning deep learning models for image classification, segmentation, and large language models (LLMs). She has been involved in contributing to high-impact AI deployments across national academic and research organizations.She has served as HPC application coordinator for PARAM Shivay(838TF), PARAM Porul(838TF) and PARAM Himalaya(838TF).
In addition to her technical achievements, Sowmya is an active educator and community mentor. She has delivered lectures on parallel programming (MPI/OpenMP) and regularly conducts hands-on national workshops focused on HPC, deep learning, and distributed AI training for premier institutions.She holds a Post Graduate Diploma in Artificial Intelligence from C-DAC ACTS Pune and a Bachelor’s degree in Information Technology from New Horizon College of Engineering, Bengaluru.
At HPCKP 2025, Sowmya will be presenting her work “Distributed Deep Learning Training Using Enroot on PARAM Rudra” . Here is a peek at her abstract:
In modern AI research, distributed training is essential to accelerate deep learning workloads. However, traditional containerization methods often introduce deployment complexity or compatibility issues on HPC clusters.In this work, Sowmya explores the use of Enroot, a lightweight container runtime optimized for NVIDIA GPU-based workloads, to run large-scale distributed training on PARAM Rudra, an A100-based supercomputing system deployed under NSM.
About Do IT Now
The Do IT Now Group has been founded to deliver a higher level of High-Performance Computing services. We are market leaders and offer the best solutions to our clients. We share a passion and enthusiasm for facing new challenges of HPC technologies together. Do IT Now deals with the complexity of HPC providing simple solutions to scientists and engineers.
The added value we offer is a deep understanding of the most advanced technologies in HPC, along with high-quality customer and user support. We offer solutions for different IT sectors like quantum computing, Big Data, artificial intelligence, cloud computing or storage.