Abstract
Containers provide flexible strategies for packing, deploying and running isolated application processes within multi-user systems and enable scientific reproducibility. The uDocker container tool offers many advantages for the development of deep learning models in the described context. The experimental results show that uDocker is more transparent to deploy for less tech-savvy researchers and allows the application to achieve processing time with negligible overhead compared to an uncontainerized environment.
Outline
💻Multi-GPU and -user systems can tackle intensive computational big data problems
🧃Containers are strategies for packing, deploying and running isolated application processes
🏍Simplify the application build and deployment process without performance penalties
👨uDocker: allows executing Docker containers without the necessity of administrative privileges
👌More transparent to deploy for less tech-savvy researchers
