Lindner, Patrick (2024) Layered Transfer: Maintaining Service Availability during Container Volume Migrations. Master's Thesis / Essay, Computing Science.
|
Text
mCS2024LindnerPM.pdf Download (658kB) | Preview |
|
![[img]](https://fse.studenttheses.ub.rug.nl/style/images/fileicons/text.png) |
Text
Toestemming.pdf Restricted to Registered users only Download (138kB) |
Abstract
Cloud computing is a great tool for delivering computing power as a utility. However, choosing for a specific cloud provider, might narrow down the deployment to the geographical locations and the servers of the cloud provider company. A migration to a different cloud provider mostly comes with great cost and downtime [15]. Technologies like Liqo provide a way of adding existing Kubernetes clusters from potentially different cloud providers to a single multi cluster, where stateless services can freely be migrated between clusters and therefore cloud providers [3]. However, the problem of live migrating stateful containers along with their attached volumes is still challenging. The aim of this thesis is to explore and improve current stateful-livecontainer-migration processes in order to provide the best quality of service and resource efficiency during a migration. The approaches are evaluated based on a collection of key performance indicators. This research focuses on two stateful container migration approaches from the literature and combines them into a novel method, seeking better performance. The combined approach is evaluated to be up to 30% less CPU consuming than the state of the art of live stateful container migration. In addition to that, for read heavy applications, the service latency can be reduced by up to 50%. This comes with the cost of more disk space consumption and for some scenarios a longer migration time. This extra disk space consumption and the additional migration time is attributed to the fact that the approach utilizes Copy-on-Write operations. However, these operations have potential to be replaced in future research.
| Item Type: | Thesis (Master's Thesis / Essay) |
|---|---|
| Supervisor name: | Lazovik, A. and Andrikopoulos, V. |
| Degree programme: | Computing Science |
| Thesis type: | Master's Thesis / Essay |
| Language: | English |
| Date Deposited: | 10 Oct 2024 10:46 |
| Last Modified: | 29 Oct 2024 06:54 |
| URI: | https://fse.studenttheses.ub.rug.nl/id/eprint/34290 |
Actions (login required)
 |
View Item |