No abstract

Cloud Computing has been used by different types of clients because it has many advantages, including the minimization of infrastructure resources costs, and its elasticity property, which allows services to be scaled up or down according to the current demand. From the Cloud provider point-of-view, there are many challenges to be overcome in order to deliver Cloud services that meet all requirements defined in Service Level Agreements (SLAs). High availability has been one of the biggest challenges for providers, and many services can be used to improve the availability of a service, such as checkpointing, load balancing, and redundancy. Beyond services, we can also find infrastructure and middleware solutions. This systematic review has as its main goal to present and discuss high available (HA) solutions for Cloud Computing, and to introduce some research challenges in this area. We hope this work can be used as a starting point to understanding and coping with HA problems in Cloud.

Abstract-This article presents a time-aware admission control and resource allocation scheme in wireless networks, in the context of a future generation cellular network. The quality levels (and their respective utility) of different connections are specified using discrete resource-utility (R-U) functions. The scheme uses these R-U functions for allocating and reallocating bandwidth to connections, aiming to maximise the accumulated utility of the system. However, different applications react differently to resource reallocations. Therefore at each allocation time point the following factors are taken into account: the age of the connection, a disconnection (drop) penalty and the sensitiveness to reallocation frequency. The evaluation of our approach shows a superior performance compared to a recent adaptive bandwidth allocation scheme (RBBS). In addition we have studied the overhead that performing a reallocation imposes on the infrastructure. To minimise this overhead, we present an algorithm that efficiently reduces the number of reallocations, while remaining within a given utility bound.

Abstract-This article proposes a scheme for bandwidth allocation in wireless ad hoc networks. The quality of service (QoS) levels for each end-to-end flow are expressed using a resourceutility function, and our algorithms aim to maximize aggregated utility. The shared channel is modeled as a bandwidth resource defined by maximal cliques of mutual interfering links.We propose a novel resource allocation algorithm that employs an auction mechanism in which flows are bidding for resources. The bids depend both on the flow's utility function and the intrinsically derived shadow prices. We then combine the admission control scheme with a utility-aware on-demand shortest path routing algorithm where shadow prices are used as a natural distance metric.As a baseline for evaluation we show that the problem can be formulated as a linear programming (LP) problem. Thus, we can compare the performance of our distributed scheme to the centralized LP solution, registering results very close to the optimum. Next we isolate the performance of price-based routing and show its advantages in hotspot scenarios, and also propose an asynchronous version that is more feasible for ad hoc environments.Further experimental evaluation compares our scheme with the state-of-the-art derived from Kelly's utility maximization framework and shows that our approach exhibits superior performance for networks with increased mobility or less frequent allocations.Index Terms-Mobile computing, pricing and resource allocation, quality of service, optimization, performance evaluation of algorithms and systems.

Abstract-This paper proposes a scheme for bandwidth allocation in wireless ad hoc networks. The Quality of Service (QoS) levels for each end-to-end flow are expressed using resource-utility functions, and our algorithms aim to maximise the aggregated utility of the flows. The scheme differentiates between applications with flexible resource requirements and rigid (real-time) requirements. As an abstract notion of resource, we use maximal cliques of mutual interfering links.Using concave piece-wise linear utility functions we present a linear programming (LP) formulation of the problem that can serve as an optimal though unrealistic solution. Then we replace this centralised approach with a distributed low complexity solution. A key concept, borrowed from the dual of the optimal allocation problem, is the shadow price of a resource.The contributions of the paper are twofold: (1) a distributed algorithm that allocates the bandwidth based on bids that are calculated using the shadow price of the resources and the flow's utility function, (2) a utility-aware on-demand "shortest" path routing algorithm in which the shadow prices are used a natural distance metric.We compare the performance of the distributed allocation scheme with the centralised, optimal linear programming solution. We also compare with a non-utility-based QoS allocation scheme, that uses hop-based shortest path routing followed by highest possible bandwidth accommodation of the flow.