A command-line tool for real-time monitoring of power consumption for CPUs, GPUs, processes, and virtual machines on GNU/Linux.
PowerJoular is a command-line tool that monitors real-time power consumption of CPUs, GPUs, individual processes, and virtual machines on GNU/Linux systems. It helps developers and researchers analyze energy usage to optimize software for efficiency and sustainability. The tool supports multiple platforms including Intel/AMD RAPL-based systems, Raspberry Pi, and Asus Tinker Board.
System administrators, software developers, and researchers focused on energy efficiency, performance optimization, and sustainability in computing environments.
PowerJoular offers low-overhead, process-level power monitoring across diverse hardware, including virtual machines, with support for continuous daemon operation and detailed CSV output for analysis.
PowerJoular allows monitoring power consumption of multiple platforms and processes
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
Monitors power on diverse systems including Intel/AMD RAPL PCs, Raspberry Pi models, and Asus Tinker Board, using appropriate interfaces for each, as listed in the supported platforms.
Identifies energy-intensive applications by monitoring individual processes in GNU/Linux with options like -p for PID or -a for app name, enabling targeted optimization.
Written in Ada and compiled to native code, ensuring low overhead during monitoring without significantly affecting system performance, as emphasized in the features.
Outputs power data to terminal and CSV files with options like -f for saving, facilitating real-time viewing and offline analysis for research or logging.
Only supports GNU/Linux, excluding Windows, macOS, and other Unix-like systems, which restricts its use in heterogeneous environments, as stated in the platform support.
Requires installing and configuring PowerJoular on both host and guest VMs with file sharing, making it cumbersome for quick VM power estimation, as detailed in the VM monitoring section.
On platforms like Raspberry Pi, power estimation relies on empirical models trained on specific revisions, potentially reducing accuracy on other hardware variations, as noted in the supported models list.