Sungjoo Kang

Model-based Autonomic Computing Framework for Cyber-Physical Systems

In this paper, we present the model-based autonomic computing framework for a cyber-physical system which provides a self-management and a self-adaptation characteristics. A development process using this framework consists of two phases: a design phase in which a developer models faults, normal status constrains, and goals of the CPS, and an operational phase in which an autonomic computing engine operates monitor-analysis-plan-execute(MAPE) cycle for managed resources of the CPS. We design a hierachical architecture for autonomic computing engines and adopt the Model Reference Adaptive Control(MRAC) as a basic feedback loop model to separate goals and resource management. According to the GroundVehicle example, we demonstrate the effectiveness of the framework.

Model Based Estimation and Tracking Method for Agile Software Project

In this paper, the authors propose a software cost estimation model for agile software development project which can help estimate concrete development costs for the desired features of a product and tracks the project progress dynamically. In general, cost estimation methods for agile developments used a story point. Because it is relative value, the estimation results tend to be easily fluctuated by the small variation of the base story point. To estimate the effort of a project development, the authors use function points in addition to the story point. The function points are determined based on the desired features of the project. The authors adopt the Kalman filter algorithm for tracking project progress. The daily variation of the function point is observed and inputted to the Kalman Filter for providing concrete estimation and velocity. The authors then validate the better performance of our model by comparing with traditional methods through a case study.

Performance evaluation of software streaming server architecture for massive users

In this paper, we propose the streaming server architecture for massive users and verify it by a small scale simulation with virtual users. Software streaming is one of the ASP solution that allows the execution of stream-enabled software, even while the transmission of the program may still be in progress. In this paper, we analyze workload characteristics of the software stream, and extract four requirements that software streaming system have to satisfy. Based on these requirements, we design the software streaming server architecture. The verification of the designed server is performed by virtual user simulation. The result tells us the capacity of each server and helps us to calculate the scale of the system for massive users

Simulation Modeling of Sewing Process for Evaluation of Production Schedule in Smart Factory

We should complete customized products in specified time to implement smart factories. For this simulation can work as a tool for validating production schedule. We propose in the manuscript a simulation model for sewing machine and modeling tool for simulation of sewing process.

CPS-based fault-tolerance method for smart factories

Abstract In this paper, we propose a fault tolerance method in smart factories which features a simulation-based framework and a fault management application, FORTUNA (Fault-tOleRanT and aUtomated decisioN making Application). The simulation-based framework is characterized by a Virtual Factory (VF) which plays an integrator role in CPS architecture for the smart factory. The VF provides two key functionalities, (1) a simulation of CPS models represented by ECML (ETRI CPS Modeling Language), and (2) a state estimation based on massive CPS data. By deploying Kalman filter, a predictor-corrector type estimator, FORTUNA not only detects faults through the realtime observation of CPS data but also predicts potential failures by the simulation. It is demonstrated that the method can work effectively in industrial systems with real dataset.

Dynamic software updating for cyber-physical systems

Cyber-Physical Systems (CPS) are highly complex systems in which the cyber and physical components are tightly integrated at all scales and level. In this paper, we propose a new dynamic software updating system for cyber-physical systems, EcoDSU. It provides a dynamic updating feature without terminating the programs and whole updating for text, partial updating for global data, and stack reconstruction for local data. We also show that EcoDSU provides some advantages over the existing systems.

Evaluation and Optimization of ActiveRecord to implement the Personalized Software Service Platform

To implement the personalized software service platform (YouFree), we use the Ruby on Rails Framework. Ruby on Rails is an excellent Web application development framework, usable for developing fast and agile web applications. It offers a number of integrated technologies and increases the productivity of web application. However, there were some weak points of ActiveRecord on Rails like limitation of find method, lazy loading problem, and so on. In this paper, we qualitatively evaluate the weak points of ActiveRecord. And we optimize the ActiveRecord to implement the YouFree. From these considerations, we can implement the more efficient YouFree system.

A Method for Distributed Personal Contents Management

This paper describes a method for managing, visualizing and searching personal contents distributed on the desktop and the multiple Web sites. Personal Web server in this system enables users to access their contents from the remote sites, so users need not upload them to the public Web server. We propose unified contents access methods for the various contents from the multiple sources and a visualization method to manage personal contents effectively. With unified contents access methods and the visualization method, all contents are accessible through the same interface and users can use them regardless of contents types or sources. To provide the up-to-date information in the personal contents management system, we implemented the file system/database synchronization engine.

Web-based Software Management System

In this paper, we propose a Web-based software management system (WebSMS) for providing the device-independent software execution environment to users. We classify software into three types based on the software execution mechanism - local software, streaming software and portable software. The proposed system merges heterogeneous software execution modules and provides Web-based interface. And it minimizes the dependencies among components and provides seamless software using interface for users.

Analysis of Software Streaming Data

In this paper, we analyze the software streaming data for stable service and efficient resource allocation of the server. We introduce the software streaming service and the message flow of the service. Based on the message flow, we firstly divide the software stream into the initial phase and the on-demand phase. And we further divide the initial phase into the downloading period and the processing period. Finally, we classify software into the downloading based software and the processing based software according to the analysis of the proportion of each period. And we also define two external environmental factors (current available network bandwidth between server and client and the processing power of the client PC) which could affect overall performance of the software streaming service. We demonstrate the external environment that affects the ALT and SRU. So we validate it by several experiments. The result shows that the streaming server consumes the most resource when the rich client uses the downloading based software in the high speed network, and there is an inverse proportion between the ALT and the SRM. Using all the information (software type, network condition, client processing power), the streaming server can allocate resource to a client more accurately.