Proactive VNF Scaling with Heterogeneous Cloud Resources: Fusing Long Short-Term Memory Prediction and Cooperative Allocation

Abstract

Network function virtualization (NFV) is designed to implement network functions by software that replaces proprietary hardware devices in traditional networks. In response to the growing demand of resource-intensive services, for NFV cloud service providers, software-oriented network functions face a number of challenges, such as dynamic deployment of virtual network functions and efficient allocation of multiple resources. This study aims at the dynamic allocation and adjustment of network multiresources and multitype flows for NFV. First, to seek a proactive approach to provision new instances for overloaded VNFs ahead of time, a model called long short-term memory recurrent neural network (LSTM RNN) is proposed to estimate flows in this paper. Then, based on the estimated flow, a cooperative and complementary resource allocation algorithm is designed to reduce resource fragmentation and improve the utilization. The final results demonstrate the advantage of the LSTM model on predicting the network function flow requirements, and our algorithm achieves good results and performance improvement in dynamically expanding network functions and improving resource utilization.

VNF-based network service consistent reconfiguration in multi-domain federations: A distributed approach

“…Proactive works predict future traffic and try to scale network services or VNFs to address these changes. While some works were proposed for dedicated services (Jia et al, 2018;Xu, 2020), we focus on shared services. For shared VNF-based network services, several works have been published.…”

“…Finally, since administrative domains have different capabilities (e.g. CPU, RAM, bandwidth) the reconfiguration operation for VNF-based network services must consider such resources to ensure functional and non-functional re-quirements (Xu, 2020). Thus, solutions for reconfiguring VNF-based network services with a focus on consistency need to be explored in the NFV context, considering both internal (hidden) and external (observable) events.…”

Improving Fairness for Distributed Interactive Applications in Software-Defined Networks

“…Hence, to figure out the controller assignment problem, the method such as DCAP in [25] can be used for data center, or scheme in [16,27] used for large scale network, but we seek a proactive approach ahead of time. erefore, we use the deep learning to estimate requests for user group based on their historical usage; the details of deep learning can be seen in our previous research [28]. Moreover, inspired by [27], but different from adjusting the facility location, we estimate the aggregate demand and adjust the control domain with the fixed number and location of controllers.…”

“…We have extracted 290, 148 network flows from the dataset and used TensorFlow to implement the models, and the python package scikit-learn to calculate performance metrics. e detail of deep learning can be seen in our previous research [28]. Same as the setting of [36], here, we set the capacity of each controller as 1800 k flows/s.…”