Resource Sharing Efficiency in Network Slicing

Abstract

The economic sustainability of future mobile networks will largely depend on the strong specialization of its offered services. Network operators will need to provide added value to their tenants, by moving from the traditional one-sizefits-all strategy to a set of virtual end-to-end instances of a common physical infrastructure, named network slices, which are especially tailored to the requirements of each application. Implementing network slicing has significant consequences in terms of resource management: service customization entails assigning to each slice fully dedicated resources, which may also be dynamically reassigned and overbooked in order to increase the cost-efficiency of the system. In this paper, we adopt a data-driven approach to quantify the efficiency of resource sharing in future sliced networks. Building on metropolitanscale real-world traffic measurements, we carry out an extensive parametric analysis that highlights how diverse performance guarantees, technological settings, and slice configurations impact the resource utilization at different levels of the infrastructure in presence of network slicing. Our results provide insights on the achievable efficiency of network slicing architectures, their dimensioning, and their interplay with resource management algorithms at different locations and reconfiguration timescales.

Statistical Assessment of IP Multimedia Subsystem in a Softwarized Environment: A Queueing Networks Approach

“…where intermediate arrival rates λ i can be derived from (14). As a general trend, Figure 4(a) reveals that, as inter-arrival times grow (corresponding in decreasing arrival rates), the mean queue length per node diminishes, as it was to be expected.…”

“…It is interesting to notice that this behavior seems to be violated by the three HSSs (in particular by HSS 1 and HSS 2 ) since they exhibit the lowest service rate (or the highest service time, according to the parameters provided in Table I). This phenomenon clearly depends on the routing probabilities that, according to (14), act as weights for λ i terms and produce the global effect of reducing the mean queue length for HSS nodes. Let us now consider the mean waiting time per node E[W i ] (accumulated across all visits) that, by applying Little's theorem to (27), can be expressed as…”

“…These processes are managed through dedicated platforms such as Docker [13], typically composed of a main engine (often referred to as the container manager) and of a certain number of instances that can be easily deployed across a different set of cloud environments. Moreover, container technology is particularly suited to implement the network slicing concepts, providing a unique opportunity to assign fully dedicated resources per slice, which can in turn be dynamically reassigned to boost the cost/efficiency trade-off of the whole system [14].…”

An Architecture for the Performance Management of Smart Healthcare Applications

“…Based on the literature, there is a diversity of network performance techniques due to unique network properties. Some applications cannot communicate their requirements, and the mapping between application requirements and network capabilities is not standard [ 38 ]. Hence, the appearance of some mechanism configurable between the applications and the network presents itself as a compelling alternative.…”

“…However, it is a crucial resource control mechanism since some applications can monopolize resources. Salvat et al [ 37 ] and Marquez et al [ 38 ] adopted a data-driven approach to quantify the efficiency of resource sharing. The authors adopted an overbooking policy, i.e., the operator allocates more resources than are actually available for the users, believing that some users will not use all resources allocated (they made an analogy with airplane and hotels companies).…”