Applications of Energy-Driven Computing

Senkans, Uvis; Balsamo, Domenico; Verykios, Theodoros D.; Merrett, Geoff V.

doi:10.1145/3142992.3142993

Cited by 7 publications

(1 citation statement)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The system state can be saved periodically, which can result in significant energy overhead for redundant checkpoints. Alternatively, snapshots can be saved on-demand when the system voltage, V sys , reaches a hibernation threshold, V h , indicating that a power outage is imminent (V sys ≤ V h ) [4], [5]. Therefore, the content needed to resume the operation from a specific point in time is stored in NVM, and forward execution can be achieved when power recovers.…”

Section: Introductionmentioning

confidence: 99%

Energy-Efficient Memory Tracing for State Retention in Transient Computing Systems

Verykios

Balsamo

Merrett

2023

2023 9th International Workshop on Advances in Sensors and Interfaces (IWASI)

Self Cite

View full text Add to dashboard Cite

Transient computing systems, also known as intermittent computing systems, are batteryless systems powered by energy harvesting (EH) sources that do not require large energy storage for system operations. Instead, they rely on retaining their state, i.e. a snapshot, in non-volatile memory (NVM) in the event of a power outage and restoring it when the power recovers. In this paper, we first discuss the limitations of stateof-the-art techniques that attempt to minimize the amount of system state saved to NVM. Therefore, we propose a novel energy-efficient system-level approach for state retention through memory tracing based on a custom hardware module named MeTra that traces changes in the main (volatile) memory between power outages. MeTra allows the voltage threshold that activates the state retention process to be dynamically adjusted according to the energy requirement of each snapshot. Thus, a great proportion of the energy harvested can be spent on useful operations. Experimental results show that the system's active time can be extended up to 17x for Flash-based systems and 92.2% for FRAM-based systems, compared to saving the entire system state, with an area overhead of as little as 2.48%.

show abstract

Section: Introductionmentioning

confidence: 99%