A probabilistic quality of service model for nondeterministic service compositions.

doctoral thesis

• Title (croatian): Vjerojatnosni model kvalitete usluge za nedeterminističke kompozicije usluga.
• Author: Adrian Satja Kurdija
• Mentor: Marin Šilić (mentor)
• Committee member: Marin Šilić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; (Department of Electronics, Microelectronics, Computer and Intelligent Systems); Zagreb
• Defense date and country: 2020, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing; Artificial Intelligence
• Universal decimal classification (UDC): 004 - Computer science and technology. Computing. Data processing

Abstract

Cloud computing becomes the prevailing aspect of software engineering. In Service-Oriented Architecture (SOA), various atomic services respond to various client requests in a cloud. Selecting the actual service instance to process a particular request can be an issue when requirements for multiple Quality of Service (QoS) properties need to be satisfied for many users simultaneously. The problem becomes more complex if we take into account the compositeness of users’ applications, which consist of many tasks with non-deterministic execution plan, where QoS properties are calculated over the whole composition. In this thesis, a fast heuristic model for multi-criteria service selection is proposed, designed for multi-user composite workflows with the goal of satisfying as many as possible of the given QoS requirements. To develop such a method, some essential tools were developed, such as probabilistic formulas for estimating QoS in non-deterministic service compositions. Another contribution is a real-time adaptive QoS prediction model, which can be used before service selection to provide its input – the QoS values for atomic services. The extensive experiments have demostrated the accuracy and efficiency of the proposed models compared to the existing ones.

Abstract (croatian)

Računarstvo u oblaku postaje prevladavajući aspekt softverskog inženjerstva. U arhitekturi zasnovanoj na uslugama, različite atomarne usluge odgovaraju na različite zahtjeve korisnika u oblaku. Odabir stvarne instance usluge za obradu određenog zahtjeva nije jednostavan kada ograničenja za različita svojstva kvalitete usluge moraju biti zadovoljena za mnoge korisnike istovremeno. Problem postaje složeniji kada uzmemo u obzir složenost korisničkih primjenskih sustava, koji se sastoje od mnogih zadataka s nedeterminističkim planom izvođenja, gdje se svojstva kvalitete usluge izračunavaju za čitavu kompoziciju. U ovoj je disertaciji predložen brzi heuristički model višekriterijskog odabira usluga, oblikovan za višekorisničke kompozicije usluga s ciljem zadovoljavanja što većeg broja zadanih ograničenja na kvalitetu usluge. U svrhu razvoja takve metode razvijeni su važni alati poput vjerojatnosnih formula za procjenu kvalitete usluge u nedeterminističkim kompozicijama. Drugi je doprinos prilagodljivi model predviđanja kvalitete usluge u stvarnom vremenu, koji se može koristiti prije odabira usluga radi dobivanja ulaznih podataka za taj odabir, tj. vrijednosti kvalitete usluge za pojedine atomarne usluge. Opsežni eksperimenti pokazali su točnost i učinkovitost predloženih modela u usporedbi s postojećim.

• Keywords: web services; quality of service; QoS; service composition; QoS prediction; service selection; collaborative filtering; transportation problem.
• Keywords (croatian): web usluge; kvaliteta usluge; kompozicija usluga; predviđanje kvalitete usluge; odabir usluga; suradničko filtriranje; problem transporta
• Language: english
• URN:NBN: urn:nbn:hr:168:436635
• Promotion: 2020
• Study programme: Title: Computer Science; Study programme type: university; Study level: postgraduate; Academic / professional title: Doktor znanosti (Doktor znanosti)
• Type of resource: Text
• Extent: 101 str.
• File origin: Born digital
• Access conditions: Open access
• Created on: 2020-09-16 15:14:00