Geoffrey Merrett

A Generic Middleware for External Peripheral State Retention in Transiently-Powered Sensor Systems

Sensor systems powered by energy harvesting usually include batteries or supercapacitors which impact the system cost and size, need time to be charged and are not environmentally friendly. In recent years, designers have proposed a new concept called transient computing that aims to remove these energy storage units and retain the systems state between power outages, in order to cope with an unreliable energy source. However, existing approaches cannot retain the state of external peripherals or are specific to certain peripherals, i.e. they are not generic. This poster proposes a generic middleware, capable to retain the state of external peripherals that are connected to a microcontroller through SPI. The validation shows the proposed approach retains the peripheral con-figuration between power failures with a maximum time overhead of 15% when configuring the peripheral. However, this represents a 0.77% overhead for a complete example application, which is lower than that caused by existing approaches.

Dataset supporting the article entitled "The Slowdown or Race-to-idle Question: Workload-Aware Energy Optimization of SMT Multicore Platforms under Process Variation"

Two widely used approaches for reducing energy consumption in multithreaded workloads are slowdown (using DVFS) and race-to-idle. In this paper, we first demonstrate that most energy-efficient choice is dependent on (1) workload (memory bound, CPU bound etc.), (2) process variation and (3) support for Simultaneous Multithreading (SMT). We then propose an approach for mapping application threads on SMT multicore systems at run-time, to minimize energy consumption. The proposed approach interfaces with the OS and hardware performance counters to characterize application threads. This characterization captures the effect of process variation on execution time and identifies the break-even operating point, where one strategy (slowdown or race-to-idle) outperforms the other. Thread mapping is performed using these characterized data by iteratively collapsing application threads (SMT) followed by binary programming-based thread mapping. Finally, performance slack is exploited at run-time to select between slowdown and race-to-idle, based upon the break-even operating point calculated for each individual thread. This end-to-end approach is implemented as a run-time manager for the Linux OS and is validated across a range of high performance applications. Results demonstrate up to 13% energy reduction over all state-of-the-art approaches, with an average of 18% improvement over Linux.

Dataset supporting the Paper titled: Intermittently-Powered Energy Harvesting Step Counter for Fitness Tracking

This Dataset supports the Paper titled Intermittently-Powered Energy Harvesting Step Counter for Fitness Tracking, accepted for publication in IEEE Sensors Applications Symposium (SAS) 2017. Funded by Mexican CONACYT.

Dataset supporting the Paper titled: Exploring ARM mbed Support for Transient Computing in Energy Harvesting IoT Systems

Energy harvesters offer the possibility for embedded IoT computing systems to operate without batteries. However, their output power is usually unpredictable and highly variable. To mitigate the effect of this variability, systems incorporate large energy buffers, increasing their size, mass and cost. The emerging class of transient computing systems differs from this approach, operating directly from the energy harvesting source and minimizing or removing additional energy storage. Different transient computing approaches have been proposed which enable computation to be sustained despite power outages. However, existing approaches are largely designed for specific applications and architectures, and hence suffer from not being broadly applicable across multiple embedded IoT platforms. To address this challenge, transient approaches need to be integrated within a general IoT programming framework such as ARMs mbed IoT Device Platform. In this paper, we explore how state-of-art transient computing approaches can be integrated into mbed, increasing ease-to-use and deployment across different platforms. This support is offered through libraries and application programming interfaces (APIs) provided by the ARM mbed OS, which enable transient computing to be implemented as a service on top of IoT application protocols. We demonstrate the ability for a transient approach to operate effectively on mbed, by practically implementing it on a low-power NXP microcontroller (MCU) with Flash memory, operating from only 1 mF additional capacitance.

Dataset supporting the article entitled “Graceful performance modulation for power neutral transient computing systems"

This dataset supports the article entitled “Graceful Performance Modulation for Power-Neutral Transient Computing Systems” accepted for publication in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems