Please use this identifier to cite or link to this item: http://hdl.handle.net/10609/146746
Title: Profiling HPC applications in containerized environments
Author: Sanuy Lostes, Albert
Tutor: Iserte, Sergio  
Others: Jorba, Josep  
Abstract: Scientific studies very often rely on supercomputers to solve difficult problems. However, reproducibility is one of the core principles in any scientific research and it is often expected that the findings of any study can be replicated with a high degree of reliability. Running these applications in containers that have been prepared and curated beforehand can provide the desired reliability, remove unnecessary complexity and reduce the manual interaction to reduce the risk of human errors. The overall goal of this project is to use Docker to create an image that contains a distributed scientific application in addition to the necessary tools that allow profiling the behaviour of the program after its execution with different workloads. The Docker containers will be managed with Singularity , the most widely used container system for HPC. The infrastructure of the HPC cluster used for this study is composed of Raspberries Pi 4 Model B hosted on-premises.
Keywords: cluster computing
profiling
scientific applications
Document type: info:eu-repo/semantics/masterThesis
Issue Date: 17-Jun-2022
Publication license: http://creativecommons.org/licenses/by-nc-nd/3.0/es/  
Appears in Collections:Trabajos finales de carrera, trabajos de investigación, etc.

Files in This Item:
File Description SizeFormat 
asanuyFMDP0622presentation.pdfPresentation of TFM4,62 MBAdobe PDFThumbnail
View/Open
asanuyFMDP0622report.pdfReport of TFM599,58 kBAdobe PDFThumbnail
View/Open
Share:
Export:
View statistics

This item is licensed under aCreative Commons License Creative Commons