Please use this identifier to cite or link to this item: http://hdl.handle.net/10609/70485
Title: High Performance Computing workload traces standardization and characterization
Author: Pérez Llamazares, Eneko
Tutor: Guim Bernat, Francesc
Others: Universitat Oberta de Catalunya
Abstract: This project aims to provide a framework for workload traces analysis and characterization. By performing such analysis, the project tries to establish a standard to define the contents of a workload trace. Besides that, the resulting framework can provide analysis results than can help understanding the data, taking decisions, etc. While analyzing a very specific data set, we concluded that three major group of variables define most of the data set. We also detected a possible anomaly with the data due to abnormal high values in two variables. Finally, a prediction was performed against this data set and we were able to see how future data should be classified.
Keywords: workload
data analysis
HPC
Document type: info:eu-repo/semantics/bachelorThesis
Issue Date: 25-Jun-2017
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 
eperezTFG2017memoria.pdf2,81 MBAdobe PDFThumbnail
View/Open
pca_google.zip2,63 kBformato zipView/Open
enekoperezTFG0118memoria.pdf2,81 MBAdobe PDFThumbnail
View/Open