Choosing the best-performing cloud for one's application is a critical problem for potential cloud customers. We propose CloudProphet, a trace-and-replay tool to predict a legacy application's performance if migrated to a cloud infrastructure. CloudProphet traces the workload of the application when running locally, and replays the same workload in the cloud for prediction. We discuss two key technical challenges in designing CloudProphet, and some preliminary results using a prototype implementation.

While many public cloud providers offer pay-as-you-go computing, their varying approaches to infrastructure, virtualization, and software services lead to a problem of plenty. To help customers pick a cloud that fits their needs, we develop CloudCmp, a systematic comparator of the performance and cost of cloud providers. CloudCmp measures the elastic computing, persistent storage, and networking services offered by a cloud along metrics that directly reflect their impact on the performance of customer applications. CloudCmp strives to ensure fairness, representativeness, and compliance of these measurements while limiting measurement cost. Applying CloudCmp to four cloud providers that together account for most of the cloud customers today, we find that their offered services vary widely in performance and costs, underscoring the need for thoughtful provider selection. From case studies on three representative cloud applications, we show that CloudCmp can guide customers in selecting the best-performing provider for their applications.

This paper presents the design and evaluation of SafeGuard, an intra-domain routing system that can safely forward packets to their destinations even when routes are changing. SafeGuard is based on the simple idea that packets carry a destination address plus a local estimate of the remaining path cost. We show that this simple design enables routers to detect path inconsistencies during route changes and resolve on a working path for anticipated failure and restoration scenarios. This in turn means that route changes do not disrupt connectivity although routing tables are inconsistent over the network. We evaluate the router performance of SafeGuard using a prototype based on NetFPGA and Quagga. We show that SafeGuard is amenable to high-speed hardware implementation with low overhead. We evaluate the network performance of SafeGuard via simulation. The results show that SafeGuard converges faster than a state-of-the-art IP fast restoration mechanism and reduces periods of disruption to a minimal duration, i.e., the failure detection time.

A Network Coordinate (NC) system is an efficient mechanism to predict Internet distance with scalable measurements. In this paper, we focus on the node churn problem – the continuous process of nodes arrival and departure – in distributed applications. Studies on Vivaldi, a representative distributed NC system, show that under node churn the prediction accuracy of the NC system will be seriously impaired. In this paper, we focus on how to handle the impact of node churn in Vivaldi. Firstly, we propose a simple solution by directly increasing the measurement frequency. Our experiments have demonstrated that this approach can reduce the harm of node churn. However, it increases the communication overhead as the measurement frequency grows. To avoid such expensive solution, we propose the design and implementation of Myth, a decentralised and fast convergence NC system. It introduces the merit of Landmark-based NC system to shorten convergence time in Vivaldi with slight extra overhead. Our experimental results show that Myth outperforms Vivaldi a lot under node churn, without compromising the performance under stable environment. Moreover, we have found that the use of Myth is a cost-effective way to achieve higher prediction accuracy; it will not only improve the prediction accuracy but also save the communication overhead.

This paper presents techniques that improve the efficiency and manageability of an IP Fast Reroute (IPFRR) technology: NotVia. NotVia provides the IPFRR service for all destinations in an ISP's network upon any single link or node failure, while previous proposals such as Loop-free Alternates (LFA) can not guarantee this level of coverage. However, NotVia increases the computational and memory costs of the IPFRR service, and poses new challenges to network management, as routers are unaware of the links and nodes (hence the amount of traffic) that they actually protect. This paper introduces three techniques: NotVia aggregation, prioritized NotVia computation, and the rNotVia algorithm that collectively reduce the overhead of NotVia and improve its manageability. We use simulations to evaluate these techniques on real ISP topologies as well as on randomly generated topologies. The results show that the computational and memory overhead of NotVia are reduced to a fraction of their previous values on various topologies, suggesting that the techniques proposed in this paper make NotVia a more efficient and easy-to-manage IPFRR solution.

As the IPv6 networks rapidly deployed, the transition problem has been changed. Because the transition of applications is a steady process with a far longer duration in comparison to the transition of infrastructure, providers cannot give up the duty of serving IPv4 traffics for the time being. On the other hand, operating a dual-stack backbone is highly costing for large-scale deployment of new networks. In this paper, a new technique of forwarding IPv4 traffics in IPv6-only backbone is developed. Rather than the ever-existing tunneling approaches, the proposed one is automatic and stateless for end nodes, without explicit tunneling. The route entries for delivering IPv4 traffics are well aggregated on the borders of IPv6 autonomous systems. This makes the technique suitable for large-scale deployment in an inter-domain networking environment.