Efficiently allocating distributed caching resources in future smart networks

Dealing with the ever-increasing video traffic is certainly one of the major challenges facing Internet Service Providers (ISPs). In this context, the strategic placement of caches is seen as one of the most important remedies, especially with recent advances in the field of virtualization. Unlike the existing works, which only focus on the placement issue, we also consider the problem of determining the optimal amount of cache to place at each possible location. We formalize, in this paper, the problem of caches placement as a multi-objective optimization problem, in which we minimize both the average distance from which contents are retrieved and the peering links utilization. As the proposed problem is NP-hard, we propose to solve it using the Greedy Randomized Adaptive Search Procedure (GRASP) meta-heuristic. Simulations results reveal the quality of the obtained solutions compared to an exhaustive search method. At the same time, they reveal that the solution is not to put all resources at the edge or at the core, as some studies claim, but to partition them judiciously, which mainly depends on the objectives of the ISPs.

show abstract

“…In this paper, we propose new extensions and substantial improvements of our previous work [10], which are summarized in the following points:…”

Section: Introductionmentioning

confidence: 99%

A GRASP-Based Approach for Dynamic Cache Resources Placement in Future Networks

Ben-Ammar

Hadjadj‐Aoul

2020

J Netw Syst Manage

Self Cite

View full text Add to dashboard Cite

show abstract

“…Mostly, state-of-the-art NFV resource orchestrators consider the static condition of networks, while ignoring the temporal differences in network traffic and topology due to mobility of users or congestion [9] [4]. Moreover, the lack of considering the recomputation of VNF placements in these methods makes them ill-equipped to be employed in practical settings, and often the consequences are violations of the Quality of Service (QoS) and Service Level Agreement (SLA) [10]. Ongoing researches evolve around different optimization formulations using Integer Linear Programming (ILP) and Mixed Integer Linear Programming (MILP) to outline the VNF orchestration scheme, which is NP-hard by nature [11] [12], and fail to offer fast VNF placements decisions at different times.…”

mentioning

confidence: 99%

“…Moreover, some researches focus solely on different VNF migration schemes [27] [28]. Ben et al have proposed a capacitated VNF migration scheme with the help of Virtual Network Embedding (VNE) [10]. However, the main focus has been deviated away from communication delay that may affect the overall user experience.…”

mentioning

confidence: 99%

Ensemble Deep Learning Assisted VNF Deployment Strategy for Next-Generation IoT Services

Emu

Choudhury

2021

IEEE Open J. Comput. Soc.

View full text Add to dashboard Cite

Due to the massive Internet of Things (IoT) connectivity and substantial growth of communication traffic, Virtual Network Function (VNF) orchestration scheme is anticipated to function promptly, dynamically, and intelligently for next-generation networks. Hence, we urge the necessity to move beyond the traditional paradigm and employ VNFs on the network edge located cloudlet. Overall, multi-access edge computing can intensify the performance of delay-sensitive IoT applications compared to the core cloud based VNF deployments. In this paper, we intend to investigate how to simultaneously leverage the ensembling of multiple deep learning models for proper calibration to provide real-time VNF placement solutions. We also address the challenges associated with state-of-the-art approaches to deal with dynamic network traffic and topology patterns. Our envisioned methods, based on Convolutional Neural Networks and Artificial Neural Networks named as E-ConvNets and E-ANN respectively, suggest two proactive VNF deployment strategies. These ensembled VNF deployment strategies demonstrate encouraging performance (optimality gap nearly 7%) in terms of minimizing relocation and communication costs, and high scalability intelligence factor (around 0.93) through simulation results compared to standalone deep learning models. Furthermore, the presented results indicate the potentialities of applying deep learning-based strategies into similar research enigmas for future telecommunication network researches.

show abstract