prikaz prve stranice dokumenta Poboljšanje raspoređivanja virtualiziranih servisa na Kubernetes platformi
Access restricted to authenticated users
master's thesis
Poboljšanje raspoređivanja virtualiziranih servisa na Kubernetes platformi
2022. urn:nbn:hr:179:656040
Cvitanović, Dina
University of Split
Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture
Department of Electronics and Computing

Cite this document

Cvitanović, D. (2022). Poboljšanje raspoređivanja virtualiziranih servisa na Kubernetes platformi (Master's thesis). Split: University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture. Retrieved from https://urn.nsk.hr/urn:nbn:hr:179:656040

Cvitanović, Dina. "Poboljšanje raspoređivanja virtualiziranih servisa na Kubernetes platformi." Master's thesis, University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, 2022. https://urn.nsk.hr/urn:nbn:hr:179:656040

Cvitanović, Dina. "Poboljšanje raspoređivanja virtualiziranih servisa na Kubernetes platformi." Master's thesis, University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, 2022. https://urn.nsk.hr/urn:nbn:hr:179:656040

Cvitanović, D. (2022). 'Poboljšanje raspoređivanja virtualiziranih servisa na Kubernetes platformi', Master's thesis, University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, accessed 24 October 2024, https://urn.nsk.hr/urn:nbn:hr:179:656040

Cvitanović D. Poboljšanje raspoređivanja virtualiziranih servisa na Kubernetes platformi [Master's thesis]. Split: University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture; 2022 [cited 2024 October 24] Available at: https://urn.nsk.hr/urn:nbn:hr:179:656040

D. Cvitanović, "Poboljšanje raspoređivanja virtualiziranih servisa na Kubernetes platformi", Master's thesis, University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, Split, 2022. Available at: https://urn.nsk.hr/urn:nbn:hr:179:656040

Please login to the repository to save this object to your list.
Metadata
TitlePoboljšanje raspoređivanja virtualiziranih servisa na Kubernetes platformi
Title (english)Improved scheduling of virtualized services on Kubernetes platform
AuthorDina Cvitanović
MentorMario Čagalj https://orcid.org/0000-0003-0235-7436 (mentor)
GranterUniversity of Split
Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture
(Department of Electronics and Computing)
Split
Defense date and country2022-07-07, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Abstract
Kubernetes platforma za orkestraciju kontejniziranih aplikacija često je korištena metoda za pokretanje cloud-native aplikacija. Ugrađena logika za raspoređivanje instanci aplikacija na klasteru (scheduler) može rezultirati stanjem u kojem se određeni broj instanci ne može pokrenuti zbog manjka resursa, a na klasteru teoretski postoji dovoljna količina slobodnih resursa. Kubernetes ne radi preraspoređivanje već pokrenutih instanci, pa ne može samostalno razriješiti takvu situaciju. Iz tog razloga, u ovom radu je razvijen i implementiran algoritam koji pomaže scheduleru tako da određene instance prerasporedi po klasteru, te omogući pokretanje dijela instanci koje su bile na čekanju. Razvijeni algoritam nije savršen, niti daje optimalan raspored, ali se tijekom testiranja u umjetno stvorenim uvjetima, kao i u promatranim realnim situacijama pokazao dovoljno dobrim i efikasnim.
Abstract (english)
Kubernetes is a platform for orchestration of containerized applications, which is common choice for running cloud-native applications. Internal logic for scheduling application instances may bring system into state where new instances cannot be started due to lack of resources, but theoretically there is enough free resources on the cluster. Kubernetes never does rescheduling of running instances, so it is not able to resolve such situations. That was a motivation for developing and implementing algorithm for rescheduling. It assists scheduler by moving some of instances within the cluster, which enables running of instances that were previously in pending state. Described algorithm is not perfect nor it results with optimal instances arrangement, but after executing the tests in artificial conditions and real-life situations, it has been concluded that it runs efficiently and gives results that are good enough.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:179:656040
Study programmeTitle: Computing
Study programme type: university
Study level: graduate
Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
Type of resourceText
File originBorn digital
Access conditionsAccess restricted to authenticated users
Terms of useLicence In copyright icon
Public noteRad izrađen u suradnji s Ericsson Nikola Tesla d.d.
Created on2023-03-10 10:53:25