Evidence record · UDINUS · SDGs
- Unit
- Rektorat UDINUS
- Period
- 2016
- Goals
- Record
- 50671cc5-1306-4e09-9ecb-396640cc86bd
- Published
- 2026-10-30
Universitas Dian Nuswantoro conducted research on a novel virtual machine selection model for dynamic virtual machine consolidation in cloud data centers. Researchers from UDINUS, namely Shidik, Guruh Fajar; Azhari; and Mustofa, Khabib, participated in the activity. The research addressed virtual machine selection as one of four sub-problems in dynamic virtual machine consolidation, which is recognized as an NP-Hard problem in the context of reducing energy consumption. The objective was to select the most suitable virtual machine for migration from an overloaded host while avoiding an oversubscribed host.
UDINUS proposed the FuMAEVMS model, based on Fuzzy Logic, Markov Normal Algorithm, and Existing VM Selection. RAM VM and maximum MIPS VM were considered as attributes. The model was evaluated using CloudSim with datasets from PlanetLabs under various conditions of virtual machine instances. Performance was measured using Energy Consumption, SLA Violation, SLA Time per active host, and Performance Degradation Due Migration. FuMAEVMS was compared with constant first selection, minimum migration time, random choice, and maximum correlation.
The findings stated that FuMAEVMS was capable of deciding which virtual machine should be migrated from an overloaded host under various virtual machine instance conditions and provided improved energy efficiency in the cloud data center. The analysis also stated that fuzzy logic categorized virtual machine attributes and helped minimize rule production in the Markov Normal Algorithm. This supported the algorithm in deciding a suitable existing virtual machine selection approach, including CFS, MMT, or MC, for migration. The experiment reported reduced energy consumption of up to 4.45% compared with RC, MMT, and MC when applied in dynamic virtual machine consolidation. The activity addressed energy consumption and energy efficiency through research on dynamic virtual machine consolidation.