Optimal Placement and Resource Provisioning for Virtual Network Functions: Traditional and Deep Reinforcement Learning Techniques
Bibliographic record
Abstract
Network Function Virtualization (NFV) marks a fundamental shift in network architecture, enabling the decoupling of network functions from proprietary hardware onto standard computational platforms, thereby enhancing network flexibility, scalability, and cost-effectiveness. This thesis explores the complex dynamics of Virtual Network Function (VNF) placement, which is paramount to realizing the full potential of NFV. The challenge of optimal VNF placement lies in balancing resource utilization, performance, and cost, which has profound implications on network efficiency and service quality. This dissertation addresses five key research problems within the domain of NFV, each pertaining to different aspects of VNF deployment and management. Firstly, it investigates strategies for optimizing VNF placement to coordinate the interplay between network resources and performance demands. This includes developing and evaluating offline and online algorithms tailored for real-world network scenarios. Secondly, the thesis tackles the issue of feasibility restoration in network optimization, proposing algorithms that can automatically detect and correct infeasibilities in VNF placement, ensuring continual service availability and compliance with network policies. Thirdly, further expanding on the theme of network optimization, this work delves into the planning of service function chains, integrating both predictive and reactive strategies to manage network resources efficiently across service providers and infrastructure providers. Applying Deep Reinforcement Learning (DRL) combined with Reward-Constrained Policy Optimization (RCPO) forms the core of the fourth study area, focusing on the dynamic optimization of online VNF placement to adapt effectively to changing network conditions and operational constraints. Lastly, the thesis introduces a novel approach to addressing resource fragmentation through a DRL-based fragmentation-aware VNF placement strategy. This method seeks to optimize the utilization of fragmented resources, enhancing overall network performance and efficiency. Collectively, the overall research is supported through theoretical analysis, algorithmic design, and extensive simulations, leading to several publications that contribute to the advancement of knowledge in network function virtualization. The findings offer practical algorithms and frameworks for enhancing VNF placement strategies and pave the way for future innovations in network management and optimization within the rapidly evolving landscape of NFV.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".