[Show abstract][Hide abstract] ABSTRACT: In large scale systems such as the Internet, replicating data is an essential feature in order to provide availability and fault-tolerance. Attiya and Welch proved that using strong consistency criteria such as atomicity is costly as each operation may need an execution time linear with the latency of the communication network. Weaker consistency criteria like causal consistency and PRAM consistency do not ensure convergence. The different replicas are not guaranteed to converge towards a unique state. Eventual consistency guarantees that all replicas eventually converge when the participants stop updating. However, it fails to fully specify the semantics of the operations on shared objects and requires additional non-intuitive and error-prone distributed specification techniques. In addition existing consistency conditions are usually defined independently from the computing entities (nodes) that manipulate the replicated data; i.e., they do not take into account how computing entities might be linked to one another, or geographically distributed. In this deliverable, we address these issues with two novel contributions. The first contribution proposes a notion of proximity graph between computing nodes. If two nodes are connected in this graph, their operations must satisfy a strong consistency condition, while the operations invoked by other nodes are allowed to satisfy a weaker condition. We use this graph to provide a generic approach to the hybridization of data consistency conditions into the same system. Based on this, we design a distributed algorithm based on this proximity graph, which combines sequential consistency and causal consistency (the resulting condition is called fisheye consistency). The second contribution of this deliverable focuses on improving the limitations of eventual consistency. To this end, we formalize a new consistency criterion, called update consistency, that requires the state of a replicated object to be consistent with a linearization of all the updates. In other words, whereas atomicity imposes a linearization of all of the operations, this criterion imposes this only on updates. Consequently some read operations may return outdated values. Update consistency is stronger than eventual consistency , so we can replace eventually consistent objects with update consistent ones in any program. Finally, we prove that update consistency is universal, in the sense that any object can be implemented under this criterion in a distributed system where any number of nodes may crash.
[Show abstract][Hide abstract] ABSTRACT: Over the last thirty years, numerous consistency conditions for replicated
data have been proposed and implemented. Popular examples of such conditions
include linearizability (or atomicity), sequential consistency, causal
consistency, and eventual consistency. These consistency conditions are usually
defined independently from the computing entities (nodes) that manipulate the
replicated data; i.e., they do not take into account how computing entities
might be linked to one another, or geographically distributed. To address this
lack, as a first contribution, this paper introduces the notion of proximity
graph between computing nodes. If two nodes are connected in this graph, their
operations must satisfy a strong consistency condition, while the operations
invoked by other nodes are allowed to satisfy a weaker condition. The second
contribution is the use of such a graph to provide a generic approach to the
hybridization of data consistency conditions into the same system. We
illustrate this approach on sequential consistency and causal consistency, and
present a model in which all data operations are causally consistent, while
operations by neighboring processes in the proximity graph are sequentially
consistent. The third contribution of the paper is the design and the proof of
a distributed algorithm based on this proximity graph, which combines
sequential consistency and causal consistency (the resulting condition is
called fisheye consistency). In doing so the paper not only extends the domain
of consistency conditions, but provides a generic provably correct solution of
direct relevance to modern georeplicated systems.
[Show abstract][Hide abstract] ABSTRACT: Inspired by the proliferation of cloud-based services, this paper studies consensus, one of the most fundamental distributed computing problems, in a hybrid model of computation. In this model, processes (or nodes) exchange information by passing messages or by accessing a reliable and highly-available register hosted in the cloud. The paper presents a formal definition of the model and problem, and studies performance tradeoffs related to using such a register. Specifically, it proves a lower bound on the number of register accesses in deterministic protocols, and gives a simple deterministic protocol that meets this bound when the register is compare-and-swap (CAS). In addition, two efficient protocols are presented; the first one is probabilistic and solves consensus with a single CAS register access in expectation, while the second one is deterministic and requires a single CAS register access when some favorable network conditions occur. A benefit of those protocols is that they can ensure both liveness and safety, and only their efficiency is affected by the probabilistic and timing assumptions.
[Show abstract][Hide abstract] ABSTRACT: Mobile cloud computing is a new rapidly growing field. In addition to the conventional fashion that mobile clients access cloud services as in the well-known client/server model, existing work has proposed to explore cloud functionalities in another perspective - offloading part of the mobile codes to the cloud for remote execution in order to optimize the application performance and energy efficiency of the mobile device. In this position paper, we investigate the state of the art of code offloading for mobile devices, highlight the significant challenges towards a more efficient cloud-based offloading framework, and also point out how existing technologies can provide us opportunities to facilitate the framework implementation.
[Show abstract][Hide abstract] ABSTRACT: This paper studies a density driven virtual topography based publish/subscribe service for mobile ad hoc networks. Two variants of the service are presented and evaluated by extensive simulations; the first is very communication frugal while the latter trades a higher message overhead for faster latency. Both variants are also being compared to two other representative approaches for publish/subscribe in ad hoc networks, a dissemination tree based scheme and an efficient flooding based scheme. It is shown that the density-driven approach outperforms the others in most tested scenarios.
[Show abstract][Hide abstract] ABSTRACT: This paper investigates a novel efficient approach to utilize multiple radio interfaces for enhancing the performance of reliable multicasts from a single sender to a group of receivers. In the proposed scheme, one radio channel (and interface) is dedicated only for recovery information transmissions. We apply this concept to both ARQ and hybrid ARQ+FEC protocols, formally analyzing the number of packets each receiver needs to process in both our approach and in the common single channel approach. We also present a corresponding efficient protocol, and study its performance by simulation. Both the formal analysis and the simulations demonstrate the benefits of our scheme.
[Show abstract][Hide abstract] ABSTRACT: Reliable broadcast is a basic service for many collaborative applications as it provides reliable dissemination of the same information to many recipients. This paper studies three common approaches for achieving scalable reliable broadcast in ad hoc networks, namely probabilistic flooding, counter-based broadcast, and lazy gossip. The strength and weaknesses of each scheme are analyzed, and a new protocol that combines these three techniques, called RAPID, is developed. Specifically, the analysis in this paper focuses on the trade-offs between reliability (percentage of nodes that receive each message), latency, and the message overhead of the protocol. Each of these methods excel in some of these parameters, but no single method wins in all of them. This motivates the need for a combined protocol that benefits from all of these methods and allows to trade between them smoothly. Interestingly, since the RAPID protocol only relies on local computations and probability, it is highly resilient to mobility and failures and even selfish behavior. By adding authentication, it can even be made malicious tolerant. Additionally, the paper includes a detailed performance evaluation by simulation. The simulations confirm that RAPID obtains higher reliability with low latency and good communication overhead compared with each of the individual methods.
Full-text · Article · Jan 2012 · IEEE Transactions on Dependable and Secure Computing
[Show abstract][Hide abstract] ABSTRACT: This paper describes a combined power and throughput performance study of WiFi and Bluetooth usage in smartphones. The work measures the obtained throughput in various settings while employing each of these technologies, and the power consumption level associated with them. In addition, the power requirements of Bluetooth and WiFi in their respective noncommunicating modes are also compared. The study reveals several interesting phenomena and tradeoffs. In particular, the paper identifies many situations in which WiFi is superior to Bluetooth, countering previous reports. The study also identifies a couple of scenarios that are better handled by Bluetooth. The conclusions from this study suggest preferred usage patterns, as well as operative suggestions for researchers and smartphone developers. This includes a cross-layer optimization for TCP/IP that could greatly improve the throughput to power ratio whenever the transmitter is more capable than the receiver.
[Show abstract][Hide abstract] ABSTRACT: Data dissemination is an important service in mobile ad hoc networks (MANETs). The main objective of this paper is to present a dissemination protocol, called locBcast, which utilizes positioning information to obtain efficient dissemination trees with low-control overhead. This paper includes an extensive simulation study that compares locBast with selfP, dominantP, fooding, and a couple of probabilistic-/counter-based protocols. It is shown that locBcast behaves similar to or better than those protocols and is especially useful in the following challenging environments: the message sizes are large, the network is dense, and nodes are highly mobile.
No preview · Article · Jan 2011 · Journal of Computer Networks and Communications
[Show abstract][Hide abstract] ABSTRACT: This paper investigates proactive data dissemination and storage schemes for wireless sensor networks (WSNs) with mobile sinks. The focus is on schemes that do not impose any restrictions on the sink’s mobility pattern. The goal is to enable the sink to collect a representative view of the network’s sensed data by visiting any set of x out of n nodes, where x≪n. The question is how to obtain this while maintaining a good trade-off between the communication overhead of the scheme, the storage space requirements on the nodes, and the ratio between the number of visited nodes x and the representativeness of the gathered data. To answer this question, we propose density-based proactivE data dissEmination Protocol (DEEP), which combines a probabilistic flooding with a probabilistic storing scheme. The DEEP protocol is formally analyzed and its performance is studied under simulations using different network densities and compared with a scheme based on random walks, called RaWMS.
Full-text · Article · May 2010 · Computer Communications
[Show abstract][Hide abstract] ABSTRACT: This paper considers the problem of calculating dominating sets in networks with bounded degree. In these networks, the maximal degree of any node is bounded by Δ, which is usually significantly smaller than n, the total number of nodes in the system. Such networks arise in various settings of wireless and peer-to-peer communication. A trivial approach of choosing all nodes into the dominating set yields an algorithm with the approximation ratio of Δ + 1. We show that any deterministic algorithm with non-trivial approximation ratio requires Ω(log*
n) rounds, meaning effectively that no o(Δ)-approximation deterministic algorithm with a running time independent of the size of the system may ever exist. On the positive side, we show two deterministic algorithms that achieve logΔ and 2logΔ-approximation in O(Δ3 + log*
n) and O(Δ2logΔ + log*
n) time, respectively. These algorithms rely on coloring rather than node IDs to break symmetry.
[Show abstract][Hide abstract] ABSTRACT: Data dissemination is an important service in ad-hoc networks. This work presents a novel approach that utilizes the positioning information and the velocity vector of the immediate neighbors at every node to extrapolate the geographic location of the surrounding nodes at any time in order to obtain efficient dissemination trees with low control overhead. The paper includes an extensive simulation study that compares our approach to other representative overlay and probabilistic forwarding approaches. It is shown that our approach behaves similar or better than the other approaches, and is especially useful in the following challenging environments: the message sizes are large, the network is dense, and nodes are highly mobile. Therefore, we believe that this protocol is appropriate for opportunistic networks.
[Show abstract][Hide abstract] ABSTRACT: Short-range wireless communication capabilities enable the creation of ad hoc networks between devices such as smart-phones or sensors, spanning, e.g., an entire high-school or a small university campus. This paper is motivated by the proliferation of devices equipped with multiple such capabilities, e.g., Blue-Tooth (BT) and WiFi for smart-phones, or ZigBee and WiFi for sensors. Yet, each of these interfaces has significantly different, and, to a large extent complementing, characteristics in terms of energy efficiency, transmission range, and bandwidth. Consequently, a viable ad hoc network composed of such devices must be able to utilize the combination of these capabilities in a clever way. For example, BT is an order of magnitude more power efficient than WiFi, but its transmission range is also an order of magnitude shorter. Hence, one would want to shut down as many WiFi transmitters as possible, while still ensuring overall network connectivity. Moreover, for latency and network capacity reasons, in addition to pure connectivity, a desired property of such a solution is to keep the number of BT hops traversed by each transmission below a given threshold k.
This paper addresses this issue by introducing the novel k-Weighted Connected Dominating Set (kWCDS) problem and providing a formal definition for it. A distributed algorithm with a proven approximation ratio is presented, followed by a heuristic protocol. While the heuristic protocol has no formally proven approximation ratio, it behaves better than the first protocol in many practical network densities. Beyond that, a tradeoff between communication overhead and the quality of the resulting kWCDS emerges. The paper includes simulation results that explore the performance of the two protocols.
[Show abstract][Hide abstract] ABSTRACT: Modern mobile phones and laptops are equipped with multiple wireless communication interfaces, such as WiFi and Bluetooth
(BT), enabling the creation of ad-hoc networks. These interfaces significantly differ from one another in their power requirements,
transmission range, bandwidth, etc. For example, BT is an order of magnitude more power efficient than WiFi, but its transmission
range is an order of magnitude shorter. This paper introduces a management middleware that establishes a power efficient overlay
for such ad-hoc networks, in which most devices can shut down their long range power hungry wireless interface (e.g., WiFi).
Yet, the resulting overlay is fully connected, and for capacity and latency needs, no message ever travels more than 2k short range (e.g., BT) hops, where k is an arbitrary parameter. The paper describes the architecture of the solution and the management protocol, as well as a
detailed simulations based performance study. The simulations largely validate the ability of the management infrastructure
to obtain considerable power savings while keeping the network connected and maintaining reasonable latency. The performance
study covers both static and mobile networks.
[Show abstract][Hide abstract] ABSTRACT: Purpose – Mobile ad-hoc networks (MANET) are networks that are formed in an ad-hoc manner by collections of devices that are equipped with wireless communication capabilities, such as the popular WiFi (IEEE 802.11b) standard. As the hardware technology and networking protocols for MANETs become mature and ubiquitous, the main barrier for MANETs to become widely used is applications. Like in other areas of distributed computing, in order to expedite the development of applications, there is a need for middleware services that support these applications. Failure detection has been identified as a basic component for many reliable distributed middleware services and applications. This paper aims to investigate this issue. Design/methodology/approach – This paper presents an adaptation of a gossip-based failure detection protocol to MANETs, and an evaluation by extensive simulations of this protocol's performance in such networks. Findings – The results can be viewed as a feasibility check for implementing failure detection in MANETs, and the conclusions drawn from them can be used to motivate and improve future implementations of both a failure detection component and of applications and middleware services relying on such a component. Originality/value – This paper presents an adaptation of a gossip-based failure detection protocol to MANET environments, and presents an extensive simulation-based performance study of this protocol in MANETs with various parameters.
Preview · Article · Nov 2009 · International Journal of Pervasive Computing and Communications
[Show abstract][Hide abstract] ABSTRACT: Hybrid networks are formed by a combination of access points and mobile nodes such that the mobile nodes can communicate both through the access points and using ad-hoc networking among themselves. This work deals with providing efficient routing between mobile devices in hybrid networks. Specifically, we assume the existence of a spanning tree from each access point to all mobile devices within the transitive transmission range of the access point. We utilize this spanning tree to design a family of efficient point-to-point routing protocols for communication between the mobile devices themselves. The protocols utilize the tree structure in order to avoid expensive flooding of the entire network. The paper includes a detailed simulation study of several representative communication patterns, which compares our approaches to DSR.
[Show abstract][Hide abstract] ABSTRACT: We propose an efficient proactive data dissemination approach that allows a mobile sink to effectively gather a representative view of a monitored region covered by n sensor nodes by visiting any m nodes, where m < < n. Moreover, the proposed strategy allows the mobile sink to follow any trajectory through the deployment region, thus decoupling the data dissemination management from the mobile sink's trajectory.
[Show abstract][Hide abstract] ABSTRACT: In this paper, we propose an adaptive and opportunistic forwarding mechanism that is able to match mobility characteristics in Delay Tolerant Networks (DTNs). Our protocol, Seeker, empowers nodes with the ability to estimate future contact opportunities based on a history of pair-wise contacts. Furthermore, Seeker is able to adjust and rewind forwarding and buffering decisions on the fly. This ensures a good trade-off between reliability and resource-efficiency, even when disconnections are frequent and hard to predict. Essentially, the novelties of Seeker are (1) its ability to estimate good relays without having a global network view and (2) its flexibility to operate under different conditions. These features are particularly important as in DTNs, devices are further restricted by strict resource constraints and a contemporaneous path may never exist between two nodes in the network. Using simulations based on a synthetic mobility model and real mobility traces, we show that Seeker is able to adapt its forwarding accordingly in diverse scenarios and achieves high performance with low overhead.