Metoda predviđanja vremena odziva SQL upita za upravljanje razinom usluga

doctoral thesis

• Title (english): SQL response time prediction method for service level management
• Author: Mario Miličević
• Mentor: Mirta Baranović (mentor)
• Committee member: Damir Kalpić (član povjerenstva)
• Committee member: Mirta Baranović (član povjerenstva)
• Committee member: Vedran Batoš (član povjerenstva)
• Committee member: Nikola Bogunović (član povjerenstva)
• Committee member: Mladen Varga (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; (Department of Applied Computing); Zagreb
• Defense date and country: 2008-07-10, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing
• Universal decimal classification (UDC): 004 - Computer science and technology. Computing. Data processing ; 005 - Management

Abstract

Radna svojstva (performanse) transakcijskih (OLTP) sustava imaju značajan utjecaj na njihovu prihvatljivost kod korisnika. Korisnički orijentiranu kvalitetu usluge (QoS) u najvećoj mjeri određuju sljedeći parametri: vrijeme odziva, raspoloživost, konzistencija i ažurnost. U ovom radu analizirana je povezanost između trenutnog opterećenja (propusnosti) poslužitelja baze podataka i vremena odziva upita. Mogućnost relativno pouzdanog predviđanja vremena odziva predstavlja temelj za automatizirani sustav odlučivanja koji može optimalno raspoređivati izvođenje upita - naročito onih zahtjevnijih koji bi mogli ugroziti dogovorenu razinu QoS.

Abstract (english)

The performances of On-Line Transaction Processing (OLTP) systems have a major effect on their acceptability to users. User oriented Quality of Service (QoS) of OLTP systems is determined with a response time, availability, consistency and currency. In this thesis the relationship between the system load/throughput and the query response time has been analyzed in the real OLTP system environment. A possibility of the accurate response time prediction may be a foundation for (automatized) decision-making for a DB query scheduling. The possible application of proposed method can be the estimation of the optimal moment for the complex query execution.

• Keywords: kvaliteta usluge; QoS; OLTP; raspoređivanje upita; strojno učenje
• Keywords (english): quality of service; QoS; OLTP; query scheduling; machine learning
• Language: croatian
• URN:NBN: urn:nbn:hr:168:815302
• Study programme: Title: Computer Science; Study programme type: university; Study level: postgraduate; Academic / professional title: Doktor znanosti (Doktor znanosti)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2020-06-01 11:01:50