2023/02/17 by Swarna Bindu Chetty, Chetty, Swarna B., Hamed Ahmadi +3
Computer Science · #Cloud Computing and Resource Management #FOS: Computer and information sciences #Networking and Internet Architecture (cs.NI) #Software System Performance and Reliability #Software-Defined Networks and 5G
paper · pdf · doi:10.48550/arxiv.2303.13523
openalex publication_date 2023/02/17 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
The Network Function Virtualization (NFV)-Resource Allocation (RA) problem is NP-Hard. Traditional deployment methods revealed the existence of a starvation problem, which the researchers failed to recognize. Basically, starvation here, means the longer waiting times and eventual rejection of low-priority services due to a 'time out'. The contribution of this work is threefold: a) explain the existence of the starvation problem in the existing methods and their drawbacks, b) introduce 'Adaptive Scheduling' (AdSch) which is an 'intelligent scheduling' scheme using a three-factor approach (priority, threshold waiting time, and reliability), which proves to be more reasonable than traditional methods solely based on priority, and c) a 'Dynamic Prioritization' (DyPr), allocation method is also proposed for unseen services and the importance of macro- and micro-level priority. We presented a zero-touch solution using Deep Deterministic Policy Gradient (DDPG) for adaptive scheduling and an online-Ridge Regression (RR) model for dynamic prioritization. The DDPG successfully identified the 'Beneficial and Starving' services, efficiently deploying twice as many low-priority services as others, reducing the starvation problem. Our online-RR model learns the pattern in less than 100 transitions, and the prediction model has an accuracy rate of more than 80%.