mruby on Resource-Constrained Low-Power Coprocessors of Embedded Devices
As IoT devices advance, their microcontroller systems-on-a-chip (SoCs) demand higher speeds, more memory, and advanced peripherals, leading to increased power consumption.
Integrating low-power (LP) coprocessors in SoCs can reduce power usage while maintaining responsiveness.
However, switching application execution to and from the coprocessors generally involves complex and platform-specific procedures.
We propose a JIT compilation method for managed programming languages to streamline LP coprocessor use.
Our prototype for the programming language mruby includes a JIT compiler and a seamless processor-switching mechanism, enabling rapid development of IoT applications leveraging LP coprocessors.
This work-in-progress paper describes the design and implementation of the extended mruby interpreter and presents preliminary evaluations of its power consumption and latency on ESP32-S3 and ESP32-C6.
| Pesentation slides (MPLR2024.pdf) | 1.14MiB |
Thu 19 SepDisplayed time zone: Amsterdam, Berlin, Bern, Rome, Stockholm, Vienna change
13:30 - 14:50 | |||
13:30 15mShort-paper | mruby on Resource-Constrained Low-Power Coprocessors of Embedded Devices MPLR A: Go Suzuki Tokyo Institute of Technology, A: Takuo Watanabe Tokyo Institute of Technology, A: Sosuke Moriguchi Tokyo Institute of Technology DOI Pre-print Media Attached File Attached | ||
13:45 15mShort-paper | Imagine There’s No Source Code: Replay Diagnostic Location Information in Dynamic EDSL Meta-programming MPLR DOI | ||
14:00 25mPaper | Existential Containers in Scala MPLR DOI | ||
14:25 25mPaper | Quff: A Dynamically Typed Hybrid Quantum-Classical Programming Language MPLR A: Christopher John Wright University of Manchester, A: Mikel Luján University of Manchester, A: Pavlos Petoumenos University of Manchester, A: John Goodacre University of Manchester DOI | ||