Presenter Profile
Arafath Nihar
PhD Student
Case Western Reserve University, Department of Computer and Data Sciences
Arafath Nihar is a Full Stack Data Scientist and doctoral candidate at Case Western Reserve University. With a focus on Spatio-Temporal Graph Convolutional Networks and expertise in data engineering, Arafath has implemented and optimized various projects, including forecasting PV time series data and developing containerized infrastructure for data science research. Their industry experience as a Software Engineer has honed their skills in application development and deployment. Arafath's strong programming proficiency, research contributions, and dedication make them a valuable asset in the field of data science.
TALK TITLE
CRADLE A Distributed and High-Performance Computing Framework for Research
KEYWORDS
Distributed computing, high-performance computing, reproducibility, usability
ABSTRACT
CRADLE (Common Research Analytics And Data Lifestyle Environment) is a groundbreaking distributed and high-performance computing framework designed to enhance research in various domains. Here we provide an overview of CRADLE, highlighting its key features, benefits, and potential impact on researc
CRADLE addresses the challenges faced in accessing computing resources for research purposes. It integrates multiple technologies and methodologies to offer a comprehensive solution. It leverages Open OnDemand, an open-source web-based platform, to provide convenient and accessible access to computing resources. It also utilizes Singularity containers which ensure secure and portable environments for high-performance computing applications. CRADLE further incorporates Hadoop, a robust big data management platform, for efficient processing and storage. Additionally, centralized software package management and FAIR (Findable, Accessible, Interoperable, and Reusable) data management principles are integrated into the framework.
The adoption of CRADLE results in significant advancements in research. Researchers utilizing CRADLE can achieve high-quality and reproducible computational science research across interdisciplinary studies. The framework facilitates efficient management of large-scale data, accelerates research productivity, and fosters scientific exploration and discovery.
In conclusion, CRADLE represents a powerful distributed and high-performance computing framework tailored for research purposes. With its integrated technologies and methodologies, CRADLE addresses resource accessibility challenges, enhances usability and reproducibility.