Published: June 7, 2024
Language: Английский
Computer Science Review, Journal Year: 2024, Volume and Issue: 52, P. 100620 - 100620
Published: Feb. 13, 2024
Language: Английский
Citations
21
Future Generation Computer Systems, Journal Year: 2023, Volume and Issue: 146, P. 222 - 233
Published: April 27, 2023
Language: Английский
Citations
16
Computing, Journal Year: 2025, Volume and Issue: 107(1)
Published: Jan. 1, 2025
Language: Английский
Citations
0
Journal of Network and Computer Applications, Journal Year: 2025, Volume and Issue: unknown, P. 104137 - 104137
Published: Feb. 1, 2025
Language: Английский
Citations
0
Mathematics, Journal Year: 2024, Volume and Issue: 12(3), P. 468 - 468
Published: Feb. 1, 2024
The advancement of cloud computing technologies has positioned virtual machine (VM) migration as a critical area research, essential for optimizing resource management, bolstering fault tolerance, and ensuring uninterrupted service delivery. This paper offers an exhaustive analysis VM processes within infrastructures, examining various types, server load assessment methods, selection strategies, ideal timing, target determination criteria. We introduce queuing theory-based model to scrutinize dynamics between servers in environment. By reinterpreting resource-centric mechanisms into task-processing paradigm, we accommodate the stochastic nature demands, characterized by random task arrivals variable processing times. is specifically tailored scenarios with two three VMs. Through numerical examples, elucidate several performance metrics: blocking probability, average tasks processed VMs, managed servers. Additionally, examine influence arrival rates duration on these measures.
Language: Английский
Citations
3
IEEE Transactions on Sustainable Computing, Journal Year: 2024, Volume and Issue: 9(5), P. 754 - 765
Published: Feb. 23, 2024
Despite continuous improvements, cloud physical resources remain underused, hence severely impacting the efficiency of these infrastructures at large.To overcome this inefficiency, Infrastructure-as-a-Service (IaaS) providers usually compensate for oversized Virtual Machines (VMs) by offering more virtual than are physically available on a host.However, technique-known as oversubscription-may hinder performances when statically-defined oversubscription ratio results in resource contention hosted VMs.Therefore, instead setting static and cluster-wide ratio, article studies how greedy increase per Physical Machine (PM) type can preserve performance goals.Keeping unchanged allows our contribution to be realistically adopted production-scale IaaS infrastructures.This contribution, named SCROOGEVM, leverages detection PM stability carefully associated ratios.Based metrics shared public providers, we investigate impact degradation.Subsequently, conduct comparative analysis SCROOGEVM with state-ofthe-art computations.The demonstrate that approach outperforms existing methods leveraging presence long-lasting VMs, while avoiding live migration penalties impacts stakeholders.
Language: Английский
Citations
3
Sustainable Computing Informatics and Systems, Journal Year: 2023, Volume and Issue: 41, P. 100948 - 100948
Published: Dec. 9, 2023
Language: Английский
Citations
9
SN Computer Science, Journal Year: 2024, Volume and Issue: 5(5)
Published: June 13, 2024
Language: Английский
Citations
2
Informatics, Journal Year: 2024, Volume and Issue: 11(3), P. 50 - 50
Published: July 19, 2024
Currently, utilizing virtualization technology in data centers often imposes an increasing burden on the host machine (HM), leading to a decline VM performance. To address this issue, live virtual migration (LVM) is employed alleviate load VM. This study introduces hybrid learning model designed estimate direct of pre-copied machines within center. The proposed integrates Markov Decision Process (MDP), genetic algorithm (GA), and random forest (RF) algorithms forecast prioritized movement identify optimal target. models achieve 99% accuracy rate with quicker training times compared previous studies that utilized K-nearest neighbor, decision tree classification, support vector machines, logistic regression, neural networks. authors recommend further exploration deep approach (DL) other center performance issues. paper outlines promising strategies for enhancing centers. demonstrate high faster than research, indicating potential optimizing placement minimizing downtime. emphasize significance considering propose investigation. Moreover, it would be beneficial delve into practical implementation dissemination real-world
Language: Английский
Citations
2
Published: July 18, 2023
Vehicular metaverses are an emerging paradigm that integrates extended reality technologies and real-time sensing data to bridge the physical space digital spaces for intelligent transportation, providing immersive experiences Metaverse Users (VMUs). VMUs access vehicular metaverse by continuously updating Twins (VTs) deployed on nearby RoadSide Units (RSUs). Due limited RSU coverage, VTs need be online migrated between RSUs ensure seamless immersion interactions with nature of mobility. However, VT migration process requires sufficient bandwidth resources from enable fast migration, leading a resource trading problem VMUs. To this end, we propose learning-based incentive mechanism task freshness-aware in metaverses. quantify freshness task, first new metric named Age Twin Migration (AoTM), which measures time elapsed completing task. Then, AoTM-based Stackelberg model, where act as leader followers. incomplete information caused privacy security concerns, utilize deep reinforcement learning learn equilibrium game. Numerical results demonstrate effectiveness our proposed
Language: Английский
Citations
6