Tram Truong-Huu

A Game-Theoretic Model for Dynamic Pricing and Competition among Cloud Providers

With many providers in todays cloud market, it is crucial for each provider to offer an optimal price policy which maximizes the final revenue and improves the competitive advantage. The competition among providers leads to the evolution of the market and dynamic resource prices over time. In this paper, we address the competition among cloud providers and propose a dynamic pricing scheme. We employ a discrete choice model to describe the users behavior. The model is used to derive the probability of a user choosing to be served by a certain provider. The competition among cloud providers is formulated as a non-cooperative stochastic game where the players are providers who act by proposing the price policy simultaneously. The game is modelled as a Markov Decision Process whose solution is a Markov Perfect Equilibrium. Numerical simulations are carried out to evaluate the performance of the proposed model.

A Stochastic Workload Distribution Approach for an Ad Hoc Mobile Cloud

Mobile devices like smartphones have become the computing device of choice for many users, heralding the era of mobile computing. Many applications have been developed to run on mobile devices. However, despite the increased processing and wireless network speeds of mobile devices, their resources are still limited in terms of processing capacity and battery lifetime. Some applications, in particular computationally intensive ones such as multimedia processing, often require more resources than a mobile device can afford. To overcome this hurdle, we propose a mobile ad-hoc cloud in which a mobile device can access resources from other sources, such as nearby mobile devices, to share the workload. The difficulty that arises with this concept is the mobility of nearby devices, i.e. A neighbouring device may move out of range before it can communicate its results back to the source node. In this paper, we propose a workload distribution scheme among these nearby mobile devices that takes into account the randomness of the connection time between cooperating devices. In order to cope with this randomness, we adopt a multi-stage stochastic programming approach which is able to take posterior recourse actions to compensate for inaccurate predictions. Numerical studies and simulations were carried out to evaluate the performance of this scheme. The results show that the stochastic programming approach outperforms a naive scheme and a baseline scheme that only considers the average connection time.

TCAM-Aware Local Rerouting for Fast and Efficient Failure Recovery in Software Defined Networks

In Software Defined Networks (SDNs), a reactive approach for failure recovery involves the centralized SDN controller which incurs long delay leading to packet losses. While a proactive approach enables fast failure recovery, it poses a new challenge concerning the number of additional forwarding rules required at every switch traversed by a flow on the primary and backup paths. These forwarding rules are stored in Ternary Content Addressable Memory (TCAM) which is limited in size and can hold only a few thousands of rules at a switch since it is expensive and power hungry. In this paper, we develop and analyze two proactive local rerouting algorithms namely Forward Local Rerouting (FLR) and Backward Local Rerouting (BLR) to compute backup paths for a primary path. By rerouting the failed traffic from the point of failure, local rerouting enables fast recovery. The proposed FLR and BLR algorithms choose backup paths so as to reduce the number of forwarding table entries with improved sharing of forwarding rules at the switches along the primary and backup paths. We evaluate the proposed algorithms through simulations on different topologies. The results show that the proposed algorithms reduce the average number of additional rules required to protect a flow by up to 75% compared to the existing approaches which do not take into account the limited size of TCAM. The results also show that the proposed algorithms are effective in terms of backup bandwidth sharing efficiency.

Dynamic embedding of workflow requests for bandwidth efficiency in data centers

In this paper, we address the problem of embedding dynamically-arriving workflow requests in data centers. Workflows pose challenges due to data precedence and time disjointness among tasks, thus driving the need for intelligent methods to embed workflows in data centers while improving the bandwidth efficiency as well as guaranteeing the application performance. We first formulate an integer programming optimization model for the embedding problem that minimizes the amount of bandwidth required for workflow execution. We then develop two algorithms namely Critical Path Workflow Embedding (CPWE) and Edge Priority Workflow Embedding (EPWE) to solve this problem. We consider two data center network architectures: packet switching electrical networks and circuit switching optical wavelength division multiplexed (WDM) networks. While WDM-based optical networks have much larger bandwidth capacity to meet the ever-growing traffic demand in data centers, they pose challenges due to wavelength continuity constraint and the nature of circuit switching. We thus additionally propose methods for selecting appropriate Top-of-the-Rack (ToR) switches and wavelengths during the embedding process so as to increase the chance of accommodating many requests that span over multiple ToRs. We evaluate CPWE and EPWE through comprehensive simulations. The results show that CPWE and EPWE significantly reduce the bandwidth required for a workflow request by up to 66% for random workflows and 80% for realistic-application workflows compared to baseline algorithms. The results also show that the proposed methods for ToR selection and wavelength selection in optical data centers outperform other methods by reducing the rejection ratio by up to 47% with dynamic reconfiguration of lightpaths and 40% with incremental configuration of lightpaths.

To Offload or to Wait: An Opportunistic Offloading Algorithm for Parallel Tasks in a Mobile Cloud

The significant development of mobile cloud computing allows a mobile user to access resources of the nearby mobile devices, i.e., Cloudlets, for processing tasks by using the offloading mechanism. However, due to the mobility of the user and cloudlets, the connection between the users device and cloudlets may be interrupted since cloudlets move out of transmission range of the users device. Consequently, the task transmission may fail, forcing the user to re-offload the task to another cloudlet or process on the local device. In this paper, we propose a dynamic opportunistic offloading algorithm which allows the user to make the decision of offloading or deferring the processing of each task in a set of parallel tasks. We formulate and solve a Markov Decision Process (MDP) model for the mobile user to obtain an optimal offloading policy while minimizing the offloading and processing cost. We extend the MDP model to a constrained MDP to solve the offloading problem when the user has a processing deadline. Numerical studies and simulations were carried out to evaluate the performance of the proposed model. The results show that the proposed model outperforms conventional baseline schemes.

Adaptive Bandwidth Allocation for Virtual Network Embedding in Optical Data Center Networks

Wavelength division multiplexed optical networks have become an attractive candidate to meet the ever-growing traffic demands in cloud data centers due to the features of large capacity and dynamic reconfiguration capability. While the bandwidth does not affect the makespan of compute-intensive and content-delivery-network applications, it has an impact on data-intensive applications that therefore require guaranteed bandwidth beside computing and storage resources for predictable performance. Motivated by this, we consider the problem of dynamically adjusting bandwidth so as to increase the acceptance of virtual networks embedded in optical data centers. We first develop an optimization programming formulation for the problem. We then develop a heuristic algorithm that efficiently embeds and adaptively allocates bandwidth to virtual networks such that the applications complete and release resources for future requests. We evaluate our algorithm through extensive simulations. The results show that our algorithm outperforms baseline algorithms by reducing rejections by up to 25%.

On Multiple Controller Mapping in Software Defined Networks With Resilience Constraints

We propose an effective switch-controller mapping scheme for distributed controller architectures in software defined networks. Our scheme maps a switch to multiple controllers and distributes flow setup requests among them to minimize flow setup time, satisfying the resilience constraint, which requires that a specified fraction of setup requests at each switch is not affected upon a controller failure. We develop an optimization formulation for the problem and compare our scheme against the single controller mapping. The results show that our scheme reduces flow setup time, provides better fairness among switches and that it is more stable against dynamic traffic fluctuations.

Handling Uncertainty and Diversity in Cloud Bandwidth Demands for Revenue Maximization

With the increasing demand for large bandwidth and diversity of bandwidth requests, to maximize the revenue, cloud providers nowadays try to offer different bandwidth request models that include guaranteed bandwidth reservation requests and on-demand flexible bandwidth requests. While guaranteed bandwidth reservation requests are beneficial for providers from the point of view of cash flow due to the upfront fee, it faces the problem of bandwidth under-utilization. On the other hand, on-demand flexible requests generate higher revenue, but they suffer from future demand uncertainty. Controlling the admission and trade-off between these kinds of requests while maximizing the revenue becomes a challenging problem for providers. In this paper, we present an optimal bandwidth allocation approach which supports the above bandwidth request models and maximizes the revenue for providers. We model the bandwidth allocation problem as a Markov Decision Process (MDP) which takes into account the utilization of guaranteed bandwidth reservation requests and the future demand uncertainty of on-demand flexible requests. We solve the MDP problem by using a dynamic programming algorithm. We demonstrate that the proposed model can be integrated into cloud data centers by leveraging on the new features of software defined networks to control the bandwidth for users. The numerical results show that the proposed model outperforms the baseline schemes and generates high revenue for providers.

SDN-based dynamic flow scheduling in optical data centers

Wavelength division multiplexing (WDM) based optical networks have been shown to be an attractive “green” solution for cloud data center networks because of its high bisection bandwidth, low power consumption and low complexity of network wiring. In addition to the circuit-switched nature of optical networks, the high degree of the optical switch and dynamic arrival of traffic flows make the problem of flow scheduling in optical data centers becomes very challenging. It has to consider not only the flow admission control but also wavelength assignment for the lightpaths. In this paper, we address the problem of flow scheduling in optical data centers. We first develop an optimization programming formulation that aims at maximizing the revenue for cloud providers in a long run. Since solving the optimization formulation is computationally prohibitive, we develop heuristic scheduling algorithms based on the congestion factor of traffic flows. We design a Software Defined Network framework that integrates the proposed algorithms to perform flow scheduling in optical data center networks. We evaluate the proposed algorithms through comprehensive simulations and compare their performance against that of a baseline algorithm. The results show that the proposed algorithms achieve good performance improvement compared to the baseline algorithm.

Fault tolerance in TCAM-limited software defined networks

In Software Defined Networks (SDNs), while a proactive fault tolerance based on the local rerouting approach enables fast failure recovery, it requires to install forwarding rules for the backup paths in the switch Ternary Content Addressable Memory (TCAM) in advance. Since the TCAM size is limited and forwarding rules are long, using large number of forwarding rules for backup paths leads to frequent eviction of flows at the switches. To guarantee the bandwidth requirement for a flow upon a link failure, bandwidth reservation is also required on the backup path. Inefficient backup bandwidth usage will cause rejection of increased number of flows. In this paper, we address the problem of proactive fault tolerance in SDNs, considering the above challenges by adopting the local rerouting approach. We develop an optimization programming formulation that determines the set of backup paths to protect a flow while minimizing the number of additional rules and bandwidth required for the backup paths. Since this problem is computationally prohibitive, we develop two heuristic algorithms, namely Forward Local Rerouting (FLR) and Backward Local Rerouting (BLR) to compute backup paths so as to improve TCAM and bandwidth usage efficiency. We propose a flexible adaptive failure recovery framework that takes the decision of using FLR or BLR based on the network state. We evaluate the proposed algorithms through simulations and compare their performance with the optimal solution. The results show that the proposed algorithms perform very close to the optimal solution and reduce the number of additional rules required to protect a flow by up to 75% compared to the approaches that do not consider the limited size of TCAM. The results also show that the proposed algorithms are effective in terms of backup bandwidth efficiency and meet the carrier-grade requirement with the recovery time below 50ms.