Dynamic task-based anycasting in mobile ad hoc networks

Mobile ad hoc networks (MANETs) have received significant attention in the recent past owing to the proliferation in the numbers of tetherless portable devices, and rapid growth in popularity of wireless networking. Most of the MANET research community has remained focused on developing lower layer mechanisms such as channel access and routing for making MANETs operational. However, little focus has been applied on higher layer issues, such as application modeling in dynamic MANET environments. In this paper, we present a novel distributed application framework based on task graphs that enables a large class of resource discovery based applications on MANETs. A distributed application is represented as a complex task comprised of smaller sub-tasks that need to be performed on different classes of computing devices with specialized roles. Execution of a particular task on a MANET involves several logical patterns of data flow between classes of such specialized devices. These data flow patterns induce dependencies between the different classes of devices that need to cooperate to execute the application. Such dependencies yield a task graph (TG) representation of the application. We focus on the problem of executing distributed tasks on a MANET by means of dynamic selection of specific devices that are needed to complete the tasks. In this paper, we present simple and efficient algorithms for dynamic discovery and selection (instantiation) of suitable devices in a MANET from among a number of them providing the same functionality. This is carried out with respect to the proposed task graph representation of the application, and we call this process Dynamic Task-Based Anycasting. Our algorithm periodically monitors the logical associations between the selected devices, and in the event of a disruption in the application owing to failures in any component in the network, it adapts to the situation and dynamically rediscovers the affected parts of the task graph, if possible. We propose metrics for evaluating the performance of these algorithms and report simulation results for a variety of application scenarios differing in complexity, traffic, and device mobility patterns. From our simulation studies, we observed that our protocol was able to instantiate and re-instantiate TG nodes quickly and yielded high effective throughput at low to medium degrees of mobility and not much below 70% effective throughput for high mobility scenarios.