ADELE: An Architecture for Steering Traffic and Computations via Deep Learning in Challenged Edge Networks
01 Oct 2019-pp 1-8
TL;DR: This paper presents an architecture for traffic management that may use deep learning to support forwarding during task offloading in challenged networked scenarios, such as those deployed by first responders after a natural or man-made disaster.
Abstract: Edge computing allows computationally intensive tasks to be offloaded to nearby (more) powerful servers, passing through an edge network. The goal of such offloading is to reduce data-intensive application response time or energy consumption, crucial constraints in mobile and IoT devices. In challenged networked scenarios, such as those deployed by first responders after a natural or man-made disaster, it is particularly difficult to achieve high levels of throughput due to scarce network conditions. In this paper, we present an architecture for traffic management that may use deep learning to support forwarding during task offloading in these challenging scenarios. In particular, our goal is to study if and when it is worth using deep learning to route traffic generated by microservices and offloading requests in these situations. Our design is different than classical approaches that use learning since we do not train for centralized routing decisions, but we let each router learn how to adapt to a lossy path without coordination, by merely using signals from standard performance-unaware protocols such as OSPF. Our results, obtained with a prototype and with simulations are encouraging, and uncover a few surprising results.
References
More filters
TL;DR: This survey paper looks at emerging research into the application of Machine Learning techniques to IP traffic classification - an inter-disciplinary blend of IP networking and data mining techniques.
Abstract: The research community has begun looking for IP traffic classification techniques that do not rely on `well known? TCP or UDP port numbers, or interpreting the contents of packet payloads. New work is emerging on the use of statistical traffic characteristics to assist in the identification and classification process. This survey paper looks at emerging research into the application of Machine Learning (ML) techniques to IP traffic classification - an inter-disciplinary blend of IP networking and data mining techniques. We provide context and motivation for the application of ML techniques to IP traffic classification, and review 18 significant works that cover the dominant period from 2004 to early 2007. These works are categorized and reviewed according to their choice of ML strategies and primary contributions to the literature. We also discuss a number of key requirements for the employment of ML-based traffic classifiers in operational IP networks, and qualitatively critique the extent to which the reviewed works meet these requirements. Open issues and challenges in the field are also discussed.
1,519 citations
07 Aug 2017
TL;DR: P Pensieve is proposed, a system that generates ABR algorithms using reinforcement learning (RL), and outperforms the best state-of-the-art scheme, with improvements in average QoE of 12%--25%.
Abstract: Client-side video players employ adaptive bitrate (ABR) algorithms to optimize user quality of experience (QoE). Despite the abundance of recently proposed schemes, state-of-the-art ABR algorithms suffer from a key limitation: they use fixed control rules based on simplified or inaccurate models of the deployment environment. As a result, existing schemes inevitably fail to achieve optimal performance across a broad set of network conditions and QoE objectives.We propose Pensieve, a system that generates ABR algorithms using reinforcement learning (RL). Pensieve trains a neural network model that selects bitrates for future video chunks based on observations collected by client video players. Pensieve does not rely on pre-programmed models or assumptions about the environment. Instead, it learns to make ABR decisions solely through observations of the resulting performance of past decisions. As a result, Pensieve automatically learns ABR algorithms that adapt to a wide range of environments and QoE metrics. We compare Pensieve to state-of-the-art ABR algorithms using trace-driven and real world experiments spanning a wide variety of network conditions, QoE metrics, and video properties. In all considered scenarios, Pensieve outperforms the best state-of-the-art scheme, with improvements in average QoE of 12%--25%. Pensieve also generalizes well, outperforming existing schemes even on networks for which it was not explicitly trained.
946 citations
TL;DR: This survey delineates the limitations, give insights, research challenges and future opportunities to advance ML in networking, and jointly presents the application of diverse ML techniques in various key areas of networking across different network technologies.
Abstract: Machine Learning (ML) has been enjoying an unprecedented surge in applications that solve problems and enable automation in diverse domains. Primarily, this is due to the explosion in the availability of data, significant improvements in ML techniques, and advancement in computing capabilities. Undoubtedly, ML has been applied to various mundane and complex problems arising in network operation and management. There are various surveys on ML for specific areas in networking or for specific network technologies. This survey is original, since it jointly presents the application of diverse ML techniques in various key areas of networking across different network technologies. In this way, readers will benefit from a comprehensive discussion on the different learning paradigms and ML techniques applied to fundamental problems in networking, including traffic prediction, routing and classification, congestion control, resource and fault management, QoS and QoE management, and network security. Furthermore, this survey delineates the limitations, give insights, research challenges and future opportunities to advance ML in networking. Therefore, this is a timely contribution of the implications of ML for networking, that is pushing the barriers of autonomic network operation and management.
677 citations
25 Aug 2003
TL;DR: It is argued that cognitive techniques, rather than traditional algorithmic approaches, are best suited to meeting the uncertainties and complexity of the objective of network research.
Abstract: We propose a new objective for network research: to build a fundamentally different sort of network that can assemble itself given high level instructions, reassemble itself as requirements change, automatically discover when something goes wrong, and automatically fix a detected problem or explain why it cannot do so.We further argue that to achieve this goal, it is not sufficient to improve incrementally on the techniques and algorithms we know today. Instead, we propose a new construct, the Knowledge Plane, a pervasive system within the network that builds and maintains high-level models of what the network is supposed to do, in order to provide services and advice to other elements of the network. The knowledge plane is novel in its reliance on the tools of AI and cognitive systems. We argue that cognitive techniques, rather than traditional algorithmic approaches, are best suited to meeting the uncertainties and complexity of our objective.
635 citations
"ADELE: An Architecture for Steering..." refers background in this paper
...To fill the performance-unaware gap of many (edge) network decision problems, the community has revived the decade old [12] idea of Data-driven networking [4], [13]– [15]....
[...]