5G represents the next generation of communication networks and services, and will bring a new set of use cases and scenarios. These in turn will address a new set of challenges from the network and service management perspective, such as network traffic and resource management, big data management and energy efficiency. Consequently, novel techniques and strategies are required to address these challenges in a smarter way. In this paper, we present the limitations of the current network and service management and describe in detail the challenges that 5G is expected to face from a management perspective. The main contribution of this paper is presenting a set of use cases and scenarios of 5G in which machine learning can aid in addressing their management challenges. It is expected that machine learning can provide a higher and more intelligent level of monitoring and management of networks and applications, improve operational efficiencies and facilitate the requirements of the future 5G network.

No abstract available.

Over the past few years Content-Centric Networking, a networking architecture in which host-to-content communication protocols are introduced, has been gaining much attention. A central component of such an architecture is a large-scale interconnected caching system. To date, the way these Cache Networks operate and perform is still poorly understood. Following the work of Cruz on queueing networks, in this paper we develop a network calculus for bounding flows in LRU cache networks of arbitrary topology. We analyze the tightness of these bounds as a function of several system parameters. Also, we derive from it several analytical results regarding these systems: the uniformizing impact of LRU on the request stream, and the significance of cache and routing diversity on performance.

Over the past few years Content-Centric Networking, a networking model in which host-to-content communication protocols are introduced, has been gaining much attention. A central component of such an architecture is a large-scale interconnected caching system. To date, the way these Cache Networks operate and perform is still poorly understood. In this work, we demonstrate that certain cache networks are non-ergodic in that their steady-state characterization depends on the initial state of the system. We then establish several important properties of cache networks, in the form of three independently-sufficient conditions for a cache network to comprise a single ergodic component. Each property targets a different aspect of the system - topology, admission control and cache replacement policies. Perhaps most importantly we demonstrate that cache replacement can be grouped into equivalence classes, such that the ergodicity (or lack-thereof) of one policy implies the same property holds for all policies in the class.

In recent years, much work has been devoted to developing protocols and architectures for supporting the growing trend of data-oriented services. One drawback of many of these proposals is the need to upgrade or replace all the routers in order for the new systems to work. Among the few systems that allow for gradual deployment is the recently-proposed Breadcrumbs technique for distributed coordination among caches in a cache network. Breadcrumbs uses information collected locally at each cache during past downloads to support in-network guiding of current requests to desired content. Specifically, during content download a series of short-term pointers, called breadcrumbs, is set up along the download path. Future requests for this content are initially routed towards the server which holds (a copy of) this content. However, if this route leads the request to a Breadcrumbs-supporting router, this router re-directs the request in the direction of the latest downloaded, using the aforementioned pointers. Thus, content requests are initially forwarded by a location ID (e.g., IP address), but encountering a breadcrumb entry can cause a shift over to content-based routing. This property enables the Breadcrumbs system to be deployed gradually, since it only enhances the existing location-based routing mechanism (i.e. IP-based routing). In this paper we evaluate the performance of a network where Breadcrumbs is only partially deployed. Our simulation results show Breadcrumbs performs poorly when sparsely deployed. However, if an overlay of Breadcrumbs-supporting routers is setup, system performance is greatly improved. We believe that the reduced load on servers achieved with even a limited deployment of Breadcrumbs-supporting routers, combined with the flexibility of being able to deploy the system gradually, should motivate further investigation and eventual deployment of Breadcrumbs.

Many systems employ caches to improve performance. While isolated caches have been studied in-depth, multi-cache systems are not well understood, especially in networks with arbitrary topologies. In order to gain insight into and manage these systems, a low-complexity algorithm for approximating their behavior is required. We propose a new algorithm, termed a-Net, that approximates the behavior of multi-cache networks by leveraging existing approximation algorithms for isolated LRU caches. We demonstrate the utility of a-Net using both per- cache and network-wide performance measures. We also perform factor analysis of the approximation error to identify system parameters that determine the precision of a-Net.

In a number of network scenarios (including military settings), mobile nodes are clustered into groups, with nodes within the same group exhibiting significant correlation in their movements. Mobility models for such networks should reflect this group structure. In this paper, we consider the problem of identifying the number of groups, and the membership of mobile nodes within groups, from a trace of mobile nodes. We present two clustering algorithms to determine the number of groups and their identities: k-means chain and spectral clustering. Different from traditional k-means clustering, k-means chain identifies the number of groups in a dynamic graph, using a chaining process to keep track of group trajectories over the entire trace. The second approach uses spectral clustering, which uses similarities between node pairs to cluster nodes into groups. We show that the number of groups and node membership can be accurately extracted from traces, particularly when the number of groups is small.

For several years, web caching has been used to meet the ever-increasing Web access loads. A fundamental capability of all such systems is that of inter-cache coordination, which can be divided into two main types: explicit and implicit coordination. While the former allows for greater control over resource allocation, the latter does not suffer from the additional communication overhead needed for coordination. In this paper, we consider a network in which each router has a local cache that caches files passing through it. By additionally storing minimal information regarding caching history, we develop a simple content caching, location, and routing systems that adopts an implicit, transparent, and best-effort approach towards caching. Though only best effort, the policy outperforms classic policies that allow explicit coordination between caches.