izpis_h1_title_alt

Avtomatska izgradnja modela odvisnosti med komponentami IT
ID Komavec, Mojca (Author), ID Bajec, Marko (Mentor) More about this mentor... This link opens in a new window, ID Helic, Denis (Comentor)

.pdfPDF - Presentation file, Download (1010,71 KB)
MD5: CF71169603C9A76148E2D3C62BE97B05
PID: 20.500.12556/rul/69b1b08a-daf9-4ef4-8a29-af083f1cfaee

Abstract
Tehnologija in infrastruktura IT postajata zelo pomembna pri velikih podjetjih, saj ponujata hiter odziv in razrešitev ponavljajočih se problemov, napak in odkritje osnovnih vzrokov le-teh. Za avtomatizacijo teh procesov se podjetja zanašajo na pristop, imenovan identifikacija izvora napak. Ena izmed komponent je modul za odkrivanje komponent, ki med drugim vsebuje tudi informacijo o odvisnostih med komponentami IT. V tej nalogi se osredotočamo na avtomatsko gradnjo grafa odvisnosti med komponentami IT iz konfiguracijskih parametrov. Podatke analiziramo tako, da najprej najdemo odvisnosti med gostitelji, nato pa med posameznimi komponentami IT. Vsaki odvisnosti z algoritmom strojnega učenja dodelimo verjetnost, da obstaja. Prav tako pokažemo, da je naš pristop hitrejši in bolj natančen kot naiven pristop, ki primerja vsak konfiguracijski parameter z vsakim. Naloga vsebuje tudi obsežno evalvacijo na resničnih podatkih. Evalvacija upošteva tranzitivno lastnost, ki velja za odvisnosti, in specifične lastnosti, ki veljajo pri analizi identifikacije izvora napak. Z rezultati evalvacije pokažemo, da naš predlagan algoritem doseže 90% priklic in 100% natančnost pri odkrivanju odvisnosti med generičnimi komponentami IT.

Language:English
Keywords:analiza podatkov, preslikava odvisnosti med komponentami, odvisnosti med konfiguracijami, analiza odvisnosti, identifikacija izvora napak, usmerjeni graf
Work type:Master's thesis/paper
Organization:FRI - Faculty of Computer and Information Science
Year:2017
PID:20.500.12556/RUL-96341 This link opens in a new window
Publication date in RUL:28.09.2017
Views:1440
Downloads:437
Metadata:XML DC-XML DC-RDF
:
Copy citation
Share:Bookmark and Share

Secondary language

Language:Slovenian
Title:Automated Generation of IT Component Dependency Models
Abstract:
Large enterprises are heavily relying on IT technology and infrastructure, which strives to quickly respond and remediate occurring problems, faults, and identify the underlying root causes. To automate this process, the enterprises rely on root cause analysis approach. One of the components is a component discovery module, which also provides information about the dependencies between IT components. In this thesis, we focus on building an IT component dependency graph from granular configuration data automatically. We analyze the configuration data in order to first infer the dependencies between hosts, and secondly, to find the dependencies between IT components. Furthermore, we assign each dependency a likelihood that it exists with a supervised machine learning algorithm. We show that our approach is much faster and accurate compared to the naive approach, which compares configuration parameters to each other. Moreover, we provide an extensive evaluation on the real dataset, where the evaluation takes into account the transitive property of dependencies, and specific properties of the root cause analysis. The evaluation results show that our proposed algorithm reaches 90% recall and 100% precision for discovering dependencies between generic IT components.

Keywords:data analysis, component dependency mapping, configuration dependencies, dependency analysis, root cause analysis, directed graph

Similar documents

Similar works from RUL:
Similar works from other Slovenian collections:

Back