We believe datacenters can benefit from more focus on per-node efficiency, performance, and predictability, versus the more common focus so far on scalability to a large number of nodes. Improving per-node efficiency decreases costs and fault recovery because fewer nodes are required for the same amount of work. We believe that the use of complex, general-purpose operating systems is a key contributing factor to these inefficiencies.

Wireless Sensor Networks (WSNs) represent a new generation of networked embedded systems that must achieve long lifetimes on scarce amounts of energy. Since radio communication accounts for the primary source of power drain in these networks, a large number of different radio power management protocols have been proposed. However, the lack of operating system support for flexibly integrating them with a diverse set of applications and network platforms has made them difficult to use. This article focuses on providing link layer support toward realizing a unified power management architecture (UPMA) for WSNs. In contrast to existing monolithic approaches, we provide (i) a set of standard interfaces that separate link layer power management protocols from common MAC level functionality, (ii) an architectural framework that allows applications to easily swap out different power-management protocols depending on its needs, and (iii) a mechanism for coordinating multiple applications with different power management requirements. We have implemented our approach on both the Mica2 and Telosb radio drivers in TinyOS-2.0, the second generation of the de facto standard operating system for WSNs. Microbenchmark results show that our approach can coordinate the power-management requirements of multiple applications in a platform independent fashion while incurring negligible overhead.

Many threads packages have been proposed for programming wireless sensor platforms. However, many sensor network operating systems still choose to provide an event-driven model, due to efficiency concerns. We present TOS-Threads, a threads package for TinyOS that combines the ease of a threaded programming model with the efficiency of an event-based kernel. TOSThreads is backwards compatible with existing TinyOS code, supports an evolvable, thread-safe kernel API, and enables flexible application development through dynamic linking and loading. In TOS-Threads, TinyOS code runs at a higher priority than application threads and all kernel operations are invoked only via message passing, never directly, ensuring thread-safety while enabling maximal concurrency. The TOSThreads package is non-invasive; it does not require any large-scale changes to existing TinyOS code.

Abstract Many wireless sensor networks must sustain long lifetimes on limited energy resources. Two major approaches, transmission power control and sleep scheduling, have been proposed to reduce the radio power consumption in the transmission state and the idle state, respectively. In this paper, we first review existing transmission power control and sleep scheduling approaches and then describe a Unified Radio Power Management framework for the design and implementation of holistic radio power management solutions in wireless sensor networks. It has two key components: (1) a novel optimization approach called Minimum Power Configuration that minimizes the aggregate radio power consumption of all ratio states and (2) a Unified Power Management Architecture (UPMA) that aims to support the flexible cross‐layer integration of different power management strategies. A novel feature of UPMA is that it enables cross‐layer coordination and joint optimization of different power management strategies that exist at multiple network layers. Copyright © 2008 John Wiley & Sons, Ltd.

Radio power management is of paramount concern in wireless sensor networks that must achieve long lifetimes on scarce amounts of energy. While a multitude of power management protocols have been proposed in the past, the lack of system support for flexibly integrating them with a diverse set of applications and network platforms has made them diffcult to use. Instead of proposing yet another power management protocol, this paper focuses on providing link layer support towards realizing a Unified Power Management Architecture (UPMA) for flexible radio power management in wireless sensor networks. In contrast to the monolithic approaches adopted by existing power management solutions, we provide (1) a set of standard interfaces that allow different power management protocols existing at the link layer to be easily implemented on top of common MAC level functionality, (2) an architectural framework for enabling these protocols to be easily swapped in and out depending on the needs of the applications that require them, and (3) a mechanism for coordinating the existence of multiple applications, each of which may have different requirements for the same underlying power management protocol. We have implemented these features on the Mica2 and Telosb radio stacks in TinyOS-2.0. Microbenchmark results demonstrate that the separation of power management from MAC level functionality incurs a negligible decrease in performance when compared to existing monolithic implementations. Two case studies show that the power management requirements of multiple applications can be easily coordinated, sometimes even resulting in better power savings than any one of them can achieve individually.

The diverse requirements of wireless sensor network applications necessitate the development of multiple media access control (MAC) protocols to meet their varying throughput, latency, and network lifetime needs. Building new MAC protocols has proven to be extremely difficult, however, given the monolithic nature of existing protocol implementations as well as their dependence on a particular radio or processor platform. To address these issues, we propose the MAC Layer Architecture (MLA), a componentbased architecture for power-efficient MAC protocol development in wireless sensor networks. MLA consists of optimized, reusable components that implement a common set of features shared by existing MAC protocols, as well as abstractions that encapsulate the intricacies of the hardware platforms they run on. Through an instantiation of MLA in TinyOS 2.0.1, we have implemented five representative MAC protocols. Empirical results show that MLA results in significant code reuse among different protocols, while achieving comparative performance and memory footprints to monolithic implementations of the same protocols.

Energy management is a critical concern in wireless sensornets. Despite its importance, sensor network operating systems today provide minimal energy management support, requiring applications to explicitly manage system power states. To address this problem, we present ICEM, a device driver architecture that enables simple, energy efficient wireless sensornet applications. The key insight behind ICEM is that the most valuable information an application can give the OS for energy management is its concurrency. Using ICEM, a low-rate sensing application requires only a single line of energy management code and has an efficiency within 1.6 % of a hand-tuned implementation. ICEM’s effectiveness questions the assumption that sensornet applications must be responsible for all power management and sensornets cannot have a standardized OS with a simple API.

No abstract available.