Giljong Yoo

Proactive self-healing system based on multi-agent technologies

Most distributed computing environments today are extremely complex and time-consuming for human administrators to manage. Thus, there is increasing demand for the self-healing and self-diagnosing of problems or errors arising in systems operating within todays ubiquitous computing environment. This paper proposes a proactive self-healing system that monitors, diagnoses and heals its own internal problems using self-awareness as contextual information. The proposed system consists of Multi-Agents that analyze the log context, error events and resource status in order to perform self-healing and self-diagnosis. To minimize the resources used by the Adapters, which monitor the logs in an existing system, we place a single process in memory. By this, we mean a single Monitoring Agent monitors the context of the logs generated by the different system components. For rapid and efficient self-healing, we use a 6-step process. The effectiveness of the proposed system is confirmed through practical experiments conducted with a prototype system.

Self-management system based on self-healing mechanism

Systems designed to be self-healing are able to heal themselves at runtime in response to changing environmental or operational circumstances. Thus, the goal is to avoid catastrophic failure through prompt execution of remedial actions. This paper proposes a self-healing mechanism that monitors, diagnoses and heals its own internal problems using self-awareness as contextual information. The self-management system that encapsulates the self-healing mechanism related to reliability improvement addresses: (1) Monitoring layer, (2) Diagnosis & Decision Layer, and (3) Adaptation Layer, in order to perform self-diagnosis and self-healing. To confirm the effectiveness of self-healing mechanism, practical experiments are conducted with a prototype system.

Fault Management for Self-Healing in Ubiquitous Sensor Network

This work concerns the development of a fault model of sensor for detecting and isolating sensor, actuator, and various faults in USNs (Ubiquitous Sensor Network). USN are developed to create relationships between humans, objects and computers in various fields. A management research of sensor nodes is very important because the ubiquitous sensor network has the numerous sensor nodes. However, Self-healing technologies are insufficient to restore when an error event occurs in a sensor node in a USN environment. A layered healing architecture for each node layer (3-tier) is needed, because most sensor devices have different capacities in USN. In this paper, we design a fault model and architecture of the sensor and sensor node separately for self-healing in USN. In order to evaluate our approach, we implement prototype of the USN fault management system to evaluate our approach. We compare the resource use of self-healing components in the general distributed computing (wired network) and the USN.

Goal-Based Automated Code Generation in Self-Adaptive System

System administrator deals with many problems, as computing environment becomes increasingly complex. Systems with an ability to recognize system states and adapt to resolve these problems offer a solution. Much experience and knowledge are required to build a self-adaptive system. Self-adaptive systems have inherent difficulties. This paper proposes a technique that automatically generates the code for the self-adaptive system. Thus the system is easier to build. Self-adaptive systems of previous research required high system resource usage. Incorrect operation could be invoked by external factors such as viruses. We propose an improved self-adaptive system approach and apply it to video conference system and robot system. We compared the lines of code, the number of classes created by the developers. We have confirmed this enhanced approach to be effective in reducing these development metrics.

Hybrid Prediction Model for improving Reliability in Self-Healing System

In ubiquitous environments, which involve an even greater number of computing devices, with more informal modes of operation, this type of problem have rather serious consequences. In order to solve these problems when they arise, effective reliable systems are required. Also, system management is changing from a conventional central administration, to autonomic computing. However, most existing research focuses on healing after a problem has already occurred. In order to solve this problem, a prediction model is required to recognize operating environments and predict error occurrence. In this paper, a hybrid prediction model through four algorithms supporting self-healing in autonomic computing is proposed. This prediction model adopts a selective healing model, according to system situations for self-diagnosing and prediction of problems using four algorithms. In this paper, a hybrid prediction model is adopted to evaluate the proposed model in a self-healing system. In addition, prediction is compared with existing research and the effectiveness is demonstrated by experiment

Proactive self-healing system for application maintenance in ubiquitous computing environment

With evolving modem IT technology, one desirable characteristic of distributed of applications is self-healing, or the ability to reconfigure themselves on the fly to circumvent failure. Thus, the goal is to avoid catastrophic failure through prompt execution of remedial actions. This paper proposes a self-healing system that monitors, diagnoses and heals its own internal problems using self-awareness as contextual information. The proposed system consists of multi agents that analyze the log context, error events and resource status in order to perform self-diagnosis and self-healing. For rapid and efficient self-healing, for developing the proposed system, we use a 6-step process: monitoring, filtering, translation, diagnosis, decision and feedback. Our experiments conducted with a prototype system confirm the effectiveness of the proposed system.

A Reconfiguration Framework for Self-Healing Software

The computing environment of today is very complex. Research that endows a system with a self-healing ability that recognizes problems arising in a target system is valuable. However, most of the existing research shows that self-healing development environments need much effort and time to analyze and model constraints. Thus, in order to these problems, this paper proposes a reconfiguration framework for self-healing software. Through these, we can reduce the efforts required for developers of self-healing systems to analyze the target system. Abnormal behavior of the target system in regard to both external and internal problems can be resolved. We use a video conference system to evaluate the proposed approach. The effectiveness of our approach, compared against existing approaches for self-healing systems is verified.

Hybrid inference architecture and model for self-healing system

Distributed computing systems are continuously increasing in complexity and cost of managing, and system management tasks require significantly higher levels of autonomic management. In distributed computing, system management is changing from a conventional central administration, to autonomic computing. However, most existing research focuses on healing after a problem has already occurred. In order to solve this problem, an inference model is required to recognize operating environments and predict error occurrence. In this paper, we proposed a hybrid inference model – ID3, Fuzzy Logic, FNN and Bayesian Network – through four algorithms supporting self-healing in autonomic computing. This inference model adopts a selective healing model, according to system situations for self-diagnosing and prediction of problems using four algorithms. Therefore, correction of error prediction becomes possible. In this paper, a hybrid inference model is adopted to evaluate the proposed model in a self-healing system. In addition, inference is compared with existing research and the effectiveness is demonstrated by experiment.