On intermittence bugs in the battery-less internet of things (WIP paper)

Abstract

The resource-constrained devices of the battery-less Internet of Things are powered off energy harvesting and compute intermittently, as energy is available. Forward progress of programs is ensured by creating persistent state. Mixed-volatile platforms are thus an asset, as they map slices of the address space onto non-volatile memory. However, these platforms also possibly introduce intermittence bugs, where intermittent and continuous executions differ. Our ongoing work on intermittence bugs includes (i) an analysis that demonstrates their presence in settings that current literature overlooks; (ii) the design of efficient testing techniques to check their presence in arbitrary code, which would be otherwise prohibitive given the sheer number of different executions to check; (iii) the implementation of an offline tool called ScEpTIC that implements these techniques. ScEpTIC finds the same bugs as a brute-force approach, but is six orders of magnitude faster. CCS Concepts • Computer systems organization → Embedded systems.

Intermittence Anomalies not Considered Harmful

“…Resuming the computation with an anomalous runtime state may lead to intermittence anomalies [11,14,18], consisting in unexpected behaviors unattainable in a continuous execution. The effects of intermittence anomalies depend on how the program interacts with the anomalous part of the runtime state [14]. For example, in Fig.…”

“…This reduces checkpoint overhead, as the system saves only the volatile slice of main memory. The use of mixed-volatile platforms may cause intermittence anomalies [11,14], due to repeated executions of non-idempotent code. Fig.…”

“…Avoiding intermittence anomalies requires to save additional checkpoints in specific program locations to break harmful sequences [14,18]. For example, in Fig.…”

NVIDIA GPGPUs Instructions Energy Consumption

“…Development tools. Prior works have highlighted the unique challenges that emerge in energy-driven, and particularly intermittent, systems [10]- [12]. Ekho [11] records IVsurfaces (current-voltage curves over time) of energy harvesters, so that they can be replayed in the lab for realistic and repeatable evaluation of energy-driven systems.…”

“…Energy-driven operation brings new research and development challenges. Being governed by energy availability, the operation of these systems becomes difficult to reason about, validate and debug [10]- [12]. Today's development, validation and debugging tools typically assume that energy is always available, rendering them impractical when targeting energy-driven operation.…”