Institutional Repository of Vilnius University
Dinaminio skaidymo elgsenos vertinimas kintančių apkrovų sąlygomis duomenų bazėse /
Abstract
dc:descriptionThis thesis examines the evaluation of dynamic sharding behavior under changing workloads. The aim of the work is to define and apply a methodology for assessing the overhead introduced during automatic balancing. The methodology is applied in an experimental MongoDB sharded cluster, where balancer-disabled and balancer-enabled scenarios are compared. The evaluation uses migration records, chunk distribution changes, query latency, throughput, CPU usage, and network I/O. The results show that balancing introduced measurable overhead in the tested environment. The overhead was most visible in P95 and P99 query latency and CPU usage, while network I/O changes are associated with chunk migration between shards. Additional load-ramp experiments did not allow a reliable general benefit threshold to be identified because of experiment limits.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2026
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Ledovskojus, Martynas,
Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- lit
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD308337790&prefLang=en_US