Toshiki Kizu

Design and implementation of scalable, transparent threads for multi-core media processor

In this paper, we propose a scalable and transparent parallelization scheme using threads for multi-core processor. The performance achieved by our scheme is scalable to the number of cores, and the application program is not affected by the actual number of cores. For the performance efficiency, we designed the threads so that they do not suspend and that they do not start their execution until the data necessary for them are available. We implemented our design using three modules: the dependency controller, which controls dependencies among threads, the thread pool, which manages the ready threads, and the thread dispatcher, which fetches threads from the pool and executes them on the core. Our design and implementation provide efficient thread scheduling with low overhead. Moreover, by hiding the actual number of cores, it realizes transparency. We confirmed the transparency and scalability of our scheme by applying it to the H.264 decoder program. With this scheme, modification of application program is not necessary even if the number of cores changes due to disparate requirements. This feature makes the developing time shorter and contributes to the reduction of the developing cost.

A near-future prediction method for low power consumption on a many-core processor

We developed a method that predicts the required number of cores for executing threads in the near future on a many-core processor. It is designed for low power consumption without performance degradation. The evaluation result confirmed the proposed method is effective on a 32-cores processor.

A novel energy-efficient data acquisition method for wearable devices

On wearable devices where an MCU primarily remains in an idle state, keeping the MCU in a deep sleep mode during idle periods can lead to significant power reduction. Since the deep sleep mode has a relatively high wake-up overhead, one of the effective techniques for power reduction is to decrease the wake-up frequency of the MCU. Meanwhile, due to the recent trend of increasing the number of sensors embedded in wearable devices, the MCU must wake up more frequently for data acquisitions from the sensors. This indicates an increase in power consumption caused by frequent wakeups. In this paper, we propose a new method to achieve power reduction of the MCU for wearable applications. The applications acquire data periodically from multiple sensors. Our proposed method achieves low power consumption by gathering the scattered data acquisitions of the sensors and decreasing the wake-up frequency. The experimental result shows that the proposed method achieved an 8.5-31% power reduction in the MCU.