KLEP: a kernel level energy profiling tool for Android: poster abstract

ABSTRACT

We propose a kernel-level energy profiling tool KLEP that can work with diverse APIs of Android. KLEP addresses the challenges of the tail energy problem and the complex interrelation between hardware components in the device energy consumption profile. KLEP collects energy-sensitive events in the kernel and measures real energy consumption of the device at the same time, and employs a LSTM neural-network-based model for energy profiling. The preliminary results show that the curves profiled by KLEP can match the actual energy consumption with low error and overhead.

References

1. Kitae Kim and et al. 2014. FEPMA: fine-grained event-driven power meter for android smartphones based on device driver layer event monitoring. DATE.
2. Abhinav Pathak and et al. 2011. Fine-Grained Power Modeling for Smartphones Using System Call Tracing. EUROSYS.
3. Chanmin Yoon and et al. 2012. AppScope: Application Energy Metering Framework for Android Smartphones using Kernel Activity Monitoring. USENIX ATC.
4. Wojciech Zaremba and et al. 2015. Recurrent Neural Network Regularization. Eprint Arxiv (2015).

Index Terms

• Computer systems organization

  • Dependable and fault-tolerant systems and networks

    • Redundancy

  • Embedded and cyber-physical systems

    • Embedded systems

    • Robotics

• Networks

  • Network properties

    • Network reliability