검색 상세

실시간성과 일치성이 보장되는 사용자 적응형 정보 제공 기법 연구

User Adaptive Information Providing Schemes Guaranteeing Real-Time and Consistency

초록/요약

Cheap processors generate many modern sensor network devices. Many services can be built upon these sensor devices. As people increasingly use many electronic devices in their home, the utility of sensor devices is even more critical. These factors impact service expectations exponentially. In ubiquitous computing environments that generate much data, to provide users with quality services, it is necessary to provide reliable aggregating and monitoring systems for various sensor devices. Service providers also need to consider the users’ preference to increase customer satisfaction and product efficacy. This dissertation researched the schemes that guarantee real-time and consistency, and provide the service for the user adaptively. Services must consider user preferences to improve user satisfaction. This dissertation approaches the problem in three main ways. First, it will be argued that real-time data aggregation and monitoring methods ought to consider sensor data properties. Another important issue is adaptive user services improving customer satisfaction. Finally consistent maintenance is important, taking into account the nature of the sensor device and the service requirements of the user. In a ubiquitous environment, there are many applications where a server disseminates information of common interest to clients and devices. For example, an advertisement server sends information from a broadcast server to display devices. An efficient information scheduling scheme for information broadcast systems is proposed to reduce average waiting time for information access while maintaining fairness between information items. The scheme allocates information items adaptively according to the relative popularity of each local server. Simulation results show that the scheme can reduce waiting times up to 30% compared to the round robin scheme while maintaining cost-effectiveness. There are many methods in maintaining consistency in the distributed computing environment. Ideally, efficient schemes for maintaining consistency should take into account the following factors: lease duration of replicated data, data access pattern and system parameters. One method used for strong consistency in the web environment is the lease method. During the proxy’s lease time from a web server, the web server can notify the modification to the proxy by invalidation or update. In this dissertation, lease protocol performance is analyzed by varying update/invalidation scheme, lease duration and read rates. Through these analyses, we can choose an adaptive lease time and proper protocol (invalidation or update scheme of the modification for each proxy in the web environment). As the number of proxies for web caching increases exponentially, a more efficient method for maintaining consistency is required. A three-tier hierarchy is proposed, on which each group and node independently and adaptively chooses the proper lease time and protocol for each proxy cache. These classifications of the scheme make proxy caching adaptive to client access patterns and system parameters. Scalable real-time monitoring is one of the important requirements in supplying smart services for customers, on time, in ubiquitous smart spaces. A scalable real-time monitoring method is proposed to aggregate large amounts of data from various sensor devices distributed over different domain areas. The real time monitoring scheme can process data from sensor devices within deadlines and is scalable based on the number of sensor devices. The scalability of monitoring can be improved by employing a hierarchical monitoring agent for sensor devices, while meeting data deadlines. Simulation results show that our real-time monitoring scheme can improve the met deadline ratio up to 27% compared to previous schemes.

more

목차

1. INTRODUCTION = 13
1.1. SCALABLE REAL-TIME MONITORING SYSTEM FOR UBIQUITOUS SMART SPACE = 19
1.2. LEASE-BASED CONSISTENCY SCHEMES IN THE WEB ENVIRONMENT = 20
1.3. EFFICIENT AND FAIR SCHEDULING FOR TWO-LEVEL INFORMATION BROADCASTING SYSTEMS = 23
2. SCALABLE REAL-TIME MONITORING SYSTEM FOR UBIQUITOUS SMART SPACE = 27
2.1. RELATED WORKS = 27
2.2. THE ARCHITECTURE OF SCALABLE REAL-TIME MONITORING SYSTEMS FOR SENSOR DEVICES = 29
2.3. SIMULATION OF MONITORING SYSTEM = 35
3. LEASE-BASED CONSISTENCY SCHEMES IN THE WEB ENVIRONMENT = 38
3.1. RELATED WORKS = 38
3.1.1. Invalidation and update = 38
3.1.2. Competitive update = 39
3.1.3. Adaptive scheme = 39
3.1.4. Lease mechanism = 39
3.1.5. Proactive DNS cache update protocol [CHEN07] = 40
3.1.6. Consistency maintenance in service discovery = 41
3.2. COST ANALYSIS FOR THE LEASE-BASED EFFICIENT NOTIFICATION SCHEME = 42
3.2.1. General case analysis of the lease-based notification scheme = 44
3.2.2. Burst case analysis of the lease-based notification scheme in the web environment = 51
3.2.3. Performance comparison = 57
3.2.4. Adaptive Scheme = 67
3.3. HIERARCHICAL CONSISTENCY MAINTENANCE = 67
3.3.1. Notification method based Consistency Hierarchy = 68
3.3.2. Lease duration-based consistency hierarchy = 72
3.3.3. Complex consistency hierarchy = 73
4. EFFICIENT AND FAIR SCHEDULING FOR TWO-LEVEL INFORMATION BROADCASTING SYSTEMS = 74
4.1. RELATED WORKS = 74
4.2. PROPOSED FAIR SCHEDULING METHOD = 76
4.3. COST EFFECTIVE FAIR SCHEDULING = 90
4.4. PERFORMANCE COMPARISONS = 92
5. CONCLUSION = 104
5.1. SCALABLE REAL-TIME MONITORING SYSTEM FOR UBIQUITOUS SMART SPACE = 104
5.2. LEASE-BASED CONSISTENCY SCHEMES IN THE WEB ENVIRONMENT = 105
5.3. EFFICIENT AND FAIR SCHEDULING FOR TWO-LEVEL INFORMATION BROADCASTING SYSTEMS = 106
5.4. CONTRIBUTION = 108
BIBLIOGRAPHY = 109

more