Please use this identifier to cite or link to this item: http://hdl.handle.net/10609/81520
Title: Predicción de errores en entornos de computación distribuida
Author: Esmorís Pena, Alberto Manuel
Tutor: Isern, David  
Others: Universitat Oberta de Catalunya
Ventura, Carles  
Abstract: Applied modern techniques (related to Artificial Intelligence, Machine Learning and Data Mining) based in statistics to build a software capable of classifying jobs running in a Data Center attending to its termination/exit status. The software takes data from a SLURM (Simple Linux Utility for Resource Management) environment and then transforms it into an adequate format (CSV) to use in combination with C5.0 algorithm, in order to extract classification rules from the data set. Obtained classification rules are used by the software to classify/predict jobs. Also, it is capable of performing certain queries that work as filters that can be applied over data. Furthermore, it has a notification system which can be configured using the aforementioned queries to consider predictions of interest.
Keywords: data processing center
artificial intelligence
prediction error
Document type: info:eu-repo/semantics/bachelorThesis
Issue Date: Jun-2018
Publication license: http://creativecommons.org/licenses/by-nc-nd/3.0/es/  
Appears in Collections:Bachelor thesis, research projects, etc.

Files in This Item:
File Description SizeFormat 
anexos.zipAnexos descritos en la memoria351,2 kBUnknownView/Open
Presentación.odpPresentación en formato ODP733,1 kBOpenDocument PresentationView/Open

Video.mp4

Vídeo explicativo del TFG116,02 MBMP4View/Open
albertoesmpTFG0618memoria.pdfMemoria del TFG524,53 kBAdobe PDFThumbnail
View/Open
albertoesmpTFG0618Ppresentación.pdfPresentación en PDF del TFG513,71 kBAdobe PDFThumbnail
View/Open