Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Improvement of Container Placement Using Multi-Objective Ant Colony Optimization
KTH, School of Electrical Engineering and Computer Science (EECS).
2019 (English)Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
Abstract [en]

High resource requirements on software containers lead to the need for cloud users to find an optimal placement for each container to maximize the resource utilization in the cloud environment. Previous methods have scheduled containers in a cloud environment, optimizing a single objective, both in theory and practice. This thesis presents a Multi-Objective Container Placement Ant Colony Optimization (MOCP-ACO) algorithm, which has not been previously researched. MOCP-ACO is a modified implementation of Ant Colony Optimization, known for solving similar optimization problems, and is compared to the Spread scheduling strategy in Docker Swarm mode. The aim of this thesis is to optimize network usage and virtual machine (VM) cost, when using replicated redundant containers, in simulation and deployed as a network heavy application in Docker Swarm mode in a cloud environment. The results show that the implemented MOCP-ACO simulations with random network traffic have a significant reduction of VM cost and total network traffic. The application-specific simulation performed better in reducing VM cost but is only slightly better in reducing network traffic. Finally, the results from the cloud environment deployment showed reduced VM cost and network usage, and that the reduced network usage resulted in improved application performance.

Abstract [sv]

Stigande resurskrav för mjukvarucontainrar i en molnmiljö har orsakat ett användarbehov av att hitta en optimal placering av containrar för att maximera resursanvändningen. Flera metoder har använts för att schemalägga containrar i en molnmiljö, men endast med fokus på att optimera ett mål, exempelvis maximerad CPU-användning. Denna uppsats presenterar en ny metod för containerschemaläggning med flera mål, en algoritm för Multi-Objective Container Placement Ant Colony Optimization (MOCP-ACO). MOCP-ACO är en modifierad version av Ant Colony Optimization, vilken är känd för att effektivt lösa liknande optimeringsproblem. Målet är att optimera nätverksanvändning och kostnaden för virtuella maskiner (VMs), genom användandet av replikerade redundanta containers. Resultaten visar att MOCP-ACO bidrar till en stor minskning av nätverkstrafik och VM-kostnad vid användning av slumpmässig nätverkstrafik, samt en stor minskning i VM-kostnad med applikationsspecifik nätverkstrafik. Utöver detta visas det hur MOCP-ACO kan användas i en molnmiljö för att minska VM-kostnad och förbättra applikationsprestanda genom minskad nätverkstrafik.

Place, publisher, year, edition, pages
2019. , p. 50
Series
TRITA-EECS-EX ; 2019:36
National Category
Computer and Information Sciences
Identifiers
URN: urn:nbn:se:kth:diva-249709OAI: oai:DiVA.org:kth-249709DiVA, id: diva2:1305653
External cooperation
Giesecke+Devrient Mobile Security
Educational program
Master of Science in Engineering - Computer Science and Technology
Supervisors
Examiners
Available from: 2019-04-30 Created: 2019-04-17 Last updated: 2019-04-30Bibliographically approved

Open Access in DiVA

fulltext(1053 kB)41 downloads
File information
File name FULLTEXT01.pdfFile size 1053 kBChecksum SHA-512
2ebf3354dbcf52c4ee4355d68a111e46fbe13b56cde4ca245e6b5ba136c62ae154a4b4a9459c175dceeb1d6e2697e4eb8bde1e3f962527a400676505e2c3a0f2
Type fulltextMimetype application/pdf

By organisation
School of Electrical Engineering and Computer Science (EECS)
Computer and Information Sciences

Search outside of DiVA

GoogleGoogle Scholar
Total: 41 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

urn-nbn

Altmetric score

urn-nbn
Total: 164 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf