EStacker: Explaining Battery-Less IoT System Performance with Energy Stacks

TL;DR

EStacker introduces a repeatable evaluation platform for battery-less IoT systems powered by energy harvesting and generates energy stacks that attribute energy consumption to hardware components and activities. It also presents ST-SP, a scaled-time and scaled-power technique that preserves energy balance while accelerating evaluations, achieving a average speed-up of $6.3\times$ with throughput error around $7.7\%$ on benchmarks. The authors demonstrate practical value through case studies: diagnosing and fixing a TOF energy issue to gain $3.3\times$ performance, and rapidly exploring a smart parking design space in about $7.7$ days (vs $41.7$ days) using ST-SP. Collectively, the work enables faster, more reliable design and optimization of battery-less EH IoT deployments, with tangible improvements in energy efficiency and design-space exploration.

Abstract

The number of Internet of Things (IoT) devices is increasing exponentially, and it is environmentally and economically unsustainable to power all these devices with batteries. The key alternative is energy harvesting, but battery-less IoT systems require extensive evaluation to demonstrate that they are sufficiently performant across the full range of expected operating conditions. IoT developers thus need an evaluation platform that (i) ensures that each evaluated application and configuration is exposed to exactly the same energy environment and events, and (ii) provides a detailed account of what the application spends the harvested energy on. We therefore developed the EStacker evaluation platform which (i) provides fair and repeatable evaluation, and (ii) generates energy stacks. Energy stacks break down the total energy consumption of an application across hardware components and application activities, thereby explaining what the application specifically uses energy on. We augment EStacker with the ST-SP optimization which, in our experiments, reduces evaluation time by 6.3x on average while retaining the temporal behavior of the battery-less IoT system (average throughput error of 7.7%) by proportionally scaling time and power. We demonstrate the utility of EStacker through two case studies. In the first case study, we use energy stack profiles to identify a performance problem that, once addressed, improves performance by 3.3x. The second case study focuses on ST-SP, and we use it to explore the design space required to dimension the harvester and energy storage sizes of a smart parking application in roughly one week (7.7 days). Without ST-SP, sweeping this design space would have taken well over one month (41.7 days).

Figures (16)

• Figure 1: Our EStacker evaluation platform for battery-less IoT systems. EStacker enables repeatable evaluation of periodic and reactive IoT applications and generates energy stacks that break down energy consumption across hardware components and application activities.
• Figure 2: The hardware architecture of a state-of-the-art battery-less IoT system colin:2018:reconfigurableju:2018:predictivenardello_camaroptera_2019afanasov:2020:batterylessruppel:2022:architectural.
• Figure 3: The EStacker evaluation platform. EStacker enables (i) repeatable evaluation of EH-based IoT systems and (ii) accurately attributing energy consumption to the IoT system components and application tasks through energy stacks.
• Figure 4: Real-time evaluation compared to 2$\times$ accelerated evaluation with ST-UP and ST-SP. Capturing both performance and temporal behavior of real-time execution requires proportionally scaling time, input, and output power.
• Figure 5: Example of task execution in a periodic IoT application. ST-SP relies on increasing the average power consumption of the application by increasing its sampling frequency.