LUNAR: A Native Table Engine for Embedded Devices

LUNAR: A Native Table Engine for Embedded Devices

Explore a 19-minute conference talk from LCTES 2023 presenting LUNAR, a novel native table storage engine for embedded devices. Discover how researchers from East China Normal University address the challenges of multiple abstraction layers in SQLite that reduce bandwidth utilization. Learn about LUNAR's cross-layer design, type-aware storage layout, and variable-size allocator that optimize RAM and I/O bandwidth usage. Understand the modular architecture that allows for on-demand module selection and optional consistency modes, tailored for resource-limited embedded systems. Gain insights into LUNAR's performance improvements over state-of-the-art approaches while consuming fewer resources, as demonstrated through experimental results.