Yung-Joon Jung

Light-Weight and Versatile Monitor for a Self-Adaptive Software Framework for IoT Systems

Today, various Internet of Things (IoT) devices and applications are being developed. Such IoT devices have different hardware (HW) and software (SW) capabilities; therefore, most applications require customization when IoT devices are changed or new applications are created. However, the applications executed on these devices are not optimized for power and performance because IoT device systems do not provide suitable static and dynamic information about fast-changing system resources and applications. Therefore, this paper proposes a light-weight and versatile monitor for a self-adaptive software framework to automatically control system resources according to the system status. The monitor helps running applications guarantee low power consumption and high performance for an optimal environment. The proposed monitor has two components: a monitoring component, which provides real-time static and dynamic information about system resources and applications, and a controlling component, which supports real-time control of system resources. For the experimental verification, we created a video transport system based on IoT devices and measured the CPU utilization by dynamic voltage and frequency scaling (DVFS) for the monitor. The results demonstrate that, for up to 50 monitored processes, the monitor shows an average CPU utilization of approximately 4% in the three DVFS modes and demonstrates maximum optimization in the Performance mode of DVFS.

Transparent and selective real-time interrupt services for performance improvement

The popularity of mobile and multimedia applications made real-time support a mandatory feature for embedded operating systems. However, the current situation is that the overall performance is significantly degraded due to the real-time support. This paper suggests a novel scheme to minimize the performance degradation in embedded operating systems with real-time support. Especially, we propose transparent and selective real-time interrupt services which transparently monitor the system and postpone interrupt handling that are not relevant to real-time tasks. The proposed scheme was implemented on the Linux 2.6 kernel and the experimental results show that our scheme improves the throughput by up to 86% for Hackbench benchmark while providing almost the same scheduling latency compared to the previous work.

A real-time responsiveness measurement method of linux-based mobile systems for P2P cloud systems

Linux-based mobile computing systems such as robots, electronic control devices, and smart-phone are the most important types of P2P cloud systems in recent days. To improve the overall performance of networked systems, each mobile computing system requires real-time characteristics. For this reason, mobile computing system developers want to know how well real-time responsiveness is supported; several real-time measurement tools have been proposed. However, those previous tools have their own measurement schemes and we think that the results from those models do not show how responsive those systems are. In this paper, we propose ELRM, a new real-time measurement method that has clear measurement interval definitions and an accurate measurement method for real-time responsiveness. We evaluate ELRM on various mobile computing systems and compare it with other existing models. As a result, our method can obtain more accurate and intuitive real-time responsiveness measurement results.

Supporting self-tuning application development for smart devices

Modern smart devices that are increasing in number and complexity of applications, have a high demand for achieving all of the desired performance requirements within a reasonable cost and time range during operation. Self-adaptive computing that enables automatic adaption of system behavior and resources in order to match the goal of application performance has been known as one of the solutions to this demand. To make self-adaptive computing possible for smart devices, we have developed a prototype self-adaptive system enabling applications to automatically adjust its performance by controlling the CPU core resources and a development tool to support the development of effective self-adaptive applications for our self-adaptive system. In this paper, we focus on the development tool. The proposed tool includes a technique for estimating optimal observing point in the code, and therefore it could greatly reduce the burden for developer to write self-adaptive applications.

Rapid development of high performance applications for heterogeneous multi-core smart devices

This paper presents an IDE for rapid development of OpenCL applications on heterogeneous processing enabled smart devices. OpenCL is an open standard for parallel programming of heterogeneous many-core found in personal computers, servers and handheld/embedded devices. The IDE for OpenCL application development presented in this paper provides not only specialized project manager, editor, builder and launcher for developing OpenCL applications run on embedded devices, but also code generator based on modeling the structure and the development methodology of OpenCL program according to OpenCL specification. We expect that it is very helpful to the developers who is not familiar with OpenCL framework and reduces developer efforts required to writing high performance application using OpenCL.

Linux Based Real-Time Feature Supporting Method and Evaluation for Information Devices

Recently, demands of information devices are increasing as we can find so many information devices around us such as smartphone, mobile internet device (MID), tablet. These characteristics of information devices services should support soft real-time based time guaranteed multimedia services and control internet appliances. In this situation, soft real-time supported system should be developed to consider as a total aspect of hardware, kernel, middleware, application. But this paper will describe soft real-time supporting and evaluation methods for information device as an aspect of only kernel.