Towards Flexible Guarantees in Clouds: Adaptive Bandwidth Allocation and Pricing

Abstract

This article focuses on the problem of bandwidth allocation to users of Cloud data centers. An interesting approach is to use advance bandwidth reservation. Such systems usually assume all requests demand either bandwidth-guarantee (BG) or timeguarantee (TG), but not both. Hence the solutions are tailored for one type of requests. A BG request demands guarantee on bandwidth; whereas a TG request demands guarantee on time for transfer of data of specified volume. We define a new model that allows users to not only submit both kinds of requests, but also specify flexible demands. We tie up the problem of bandwidth allocation with differential pricing, that gives discounts to users based on the flexibility in their requests. We propose a two-phase, adaptive and flexible bandwidth allocator (A-FBA) that, in one phase admits and allocates minimal bandwidth to dynamically arriving user requests, and in another phase, allocates additional bandwidth for accepted requests maximizing revenue. The problem formulated in first phase is N P-hard, while the second phase can be solved in polynomial time. We show that, in comparison to a traditional deterministic model, the A-FBA not only increases the number of accepted requests significantly, but also does so by generating higher revenues.

Joint Optimization of Resource Provisioning in Cloud Computing

“…Bandwidth demand is a random variable per VM, which is a similar model of bandwidth demand to the one we employ here. [28] allows advance bandwidth reservation for users connecting to cloud data centers. Allocation is in two phases, firstly responding minimally to dynamically arriving requests, and secondly allocating additional bandwidth as required.…”

Fairly Scheduled Bandwidth for the Data Centre Traffic

“…If BG is selected, the bandwidth is guaranteed. TG provides an assurance of data transfer of the intended data size [14].…”

QoS-Based Rate Control Scheme for Non-Elastic Traffics in Distributed Networks

“…If RT T i and ρ are obtained, flow i can achieve its target throughput B i by selecting appropriate values for α i and β i according to (1). RT T i can be estimated in the same way as performed in TCP, i.e., the source times how long it takes for each packet sent over the channel to be acknowledged, producing a sequence of RTT samples, and then averages those samples into a "smoothed" RTT estimate.…”

“…QoS-based bandwidth allocation can be easily realized in a centralized manner for networks with central controllers. The central controller allocates the specified bandwidth for flows based on their requests [1]. Since the centralized manner needs a central node to implement bandwidth allocation, it is not suitable for distributed networks which do not have central control points, e.g., aeronautical ad hoc networks [2].…”