Section: New Results

Lightweight Streaming Middleware for the Internet of Things

Participants : Benjamin Billet, Valérie Issarny.

The Internet of Things (IoT) is a promising concept toward pervasive computing as it may radically change the way people interact with the physical world. One of the challenges raised by the IoT is the in-network continuous processing of data streams presented by Things, which must be investigated urgently because it affects the future data models of the IoT. This cross-cutting concern has been previously studied in the context of Wireless Sensor and Actuator Networks (WSAN) given the focus on the acquisition and in-network processing of sensed data. However, proposed solutions feature heterogeneous technologies that are difficult to integrate and complex to use, which represents a hurdle to their wide deployment. In addition, new types of smart sensors are emerging due to technological advances (e.g., Oracle SunSpot), enabling the implementation of complex processing tasks directly into the network, without using proxies or sending every data to the cloud. There is thus a need for a distributed middleware solution for data stream management that leverages existing WSAN work, while integrating it with today's Web technologies in order to improve the flexibility and the interoperability of the future IoT. Toward that goal, we have been developing Dioptase, a Data Stream Management System for the IoT, which aims to integrate the Things and their streams into today's Web by presenting sensors and actuators as services. The middleware specifically provides a way to describe complex fully-distributed stream-based mashups and to deploy them dynamically, at any time, as task graphs, over available Things of the network, including resource-constrained ones. To this end, Dioptase enables task graphs to be composed of Thing-specific tasks (directly implemented on the Thing) and dynamic tasks that communicates using data streams. Dynamic tasks are then described in a lightweight DSL, which is directly interpreted by the middleware and provides specific primitives to manipulate data streams.

As part of the design of Dioptase, we have been investigating dedicated task mapping. Task mapping, which basically consists of mapping a set of tasks onto a set of nodes, is a well-known problem in distributed computing research. However, as a particular case of distributed systems, the Internet of Things (IoT) poses a set of renewed challenges, because of its scale, heterogeneity and properties traditionally associated with WSAN, shared sensing, continous processing of data streams and real time computing. To handle IoT features, we present a formalization of the task mapping problem that captures the varying consumption of resources and various constraints (location, capabilities, QoS) in order to compute a mapping that guarantees the lifetime of the concurrent tasks inside the network and the fair allocation of tasks among the nodes (load balancing). It results in a binary programming problem for which we provide an efficient heuristic that allows its resolution in polynomial time. Our experiments show that our heuristic: (i) gives solutions that are close to optimal and (ii) can be implemented on reasonably powerful Things and performed directly within the network, without requiring any centralized infrastructure.