Ssu-Hsuan Lu

Pervasive health service system: insights on the development of a grid-based personal health service system

Although medical technologies developed in the twenty-first century have successfully increased mans life span, the pressure of modern life has consequently brought many modern civilization diseases and chronic illness. When all these problems are tackled by hospitals, they will consume considerable amount of medical resources. Alternatively, providing health care services at home is an important issue for improving personal health and save hospital resources. In this paper, we present an ongoing project that designs and implements a pervasive health service infrastructure based on the grid system which is integrated with the P2Ps resource sharing mechanism, to provide the personal health service. The personal health status is recorded, monitored, and even mined in/from the proposed pervasive health service system for preventive medicine. Additionally, wireless sensor equipments for mobile personal health services are also integrated into the pervasive health service system, in order to construct a situation-aware, context-aware and environment-aware mobile-health-service platform.

On the Design and Implementation of an Effective Prefetch Strategy for DSM Systems

Distributed Shared Memory systems provide the abstraction of a shared address space among computing hosts interconnected via a private network, in a convenient and easy way to achieve high performance. However, there are several drawbacks to these systems, such as communication overhead, network latency, false sharing, coherence and page faults. By prefetching, one can overlap communication and computation though the Accumulated Waiting Phenomenon and the Waiting Synchronization Phenomenon affect overall performance. In this paper, an Effective Prefetch Strategy (EPS) is proposed, to improve the shortcomings of previous prefetching approaches and increases the prefetch page hit rate. In addition, our EPS strategy reduces the waiting time for each computing host upon barrier synchronizations and misprefetches. Experimental comparisons show that our proposed EPS strategy offers the best performance among existinrefetching strategies.

Design and analysis of arrangement graph-based overlay systems for information sharing

With the continuous innovation of advanced Internet technology, Peer-to-Peer (P2P) system emerges as an important information-sharing system for the widespread exchange of resources and information among thousands of users. In this study, we applied properties of arrangement graphs to design a newly structured overlay system, named as Arrangement Graph-based Overlay (AGO). In such an overlay, the IDs between the two adjacent nodes differ only one digit and thus, the joining and leaving processes are easy while keeping the low maintenance cost. Furthermore, the searching performance in the AGO system is efficient, adaptive and scalable. Analyses of experimental results show that the establishment of the system and the node searching could achieve better performance than those obtained in the Chord system.

On design of agent home scheme for prefetching strategy in DSM systems

In distributed shared memory (DSM) systems, it is the common need to access data in remote nodes. Thus, it induces to remote access performance latencies, which is the major factor of overhead for DSM systems. Prefetching strategies can improve these phenomena by reducing latencies, but it adds workload to home nodes. It is proposed in this paper a method to reduce overhead of home nodes, by providing an agent home to share the workload of home nodes, by distributing these workloads to other nodes, when sending data. The performance evaluation of proposed strategy is done by performing three well-known benchmark programs: NPB/IS, 3DFFT and Red-Black SOR. The experimental results show that our proposed agent home method achieves 8%- 40% of speedup against original JIAJIA.

A scalable P2P overlay based on arrangement graph with minimized overhead

With innovations in the Internet, it is becoming increasingly relied upon. In the last decade, research on peer-to-peer (P2P) technology has become even more popular. As more people use P2P systems, the scalability and flexibility of the systems must be considered. In this study, an arrangement graph is used to form a P2P overlay, the Arrangement-Graph Overlay (AGO), to reduce system overhead and bind routing hops. The proposed AGO utilizes the properties of the arrangement graph, i.e., that each node has a unique ID and IDs between adjacent nodes differ by only one digit, to form the overlay network and develop a routing algorithm. The routing hops of the proposed AGO system can be bound within a certain number because of the diameter of the arrangement graph. Experimental results show that the proposed AGO system can greatly reduce system overhead and perform routing in a constant number of hops, even in a large-scale network environment. The experimental results also show that the AGO system consumes less bandwidth, which is an important consideration in P2P systems.

Effectiveness of a replica mechanism to improve availability with Arrangement Graph-Based Overlay

Peer-to-peer (P2P) overlay networks continue to evolve and grow to meet the challenges of a new age. Because peers can join or depart overlay networks at any time, researchers are particularly interested in how peers should be allowed to join overlay networks as well as how to minimize overhead in overlay networks. An Arrangement Graph-based Overlay (AGO) system can efficiently reduce system overhead by reducing the number of messages in a large-scale environment. However, AGO produces too many polling messages for the joining process. To address this issue, an enhanced joining strategy greatly improves the joining process of AGO to reduce the number of joining messages in this paper which is called the Enhanced Arrangement Graph-based Overlay (EAGO). Besides, for the purpose of the effectiveness of resource prefetching, a replica mechanism is employed to further improve routing performance. Experimental results indicate that EAGO reduces the number of joining messages by at least 20% compared with AGO, and it reduces the average number of routing hops due to the replica mechanism.

Arrangement Graph-Based Overlay with Replica Mechanism for File Sharing

Over the past decade, the development of Internet technology has led to more awareness on the powerfulness of Peer-to-Peer (P2P) overlay network. How to efficiently establish or maintain overlay networks in large-scale environments always are important issues. Besides, the ways of improving routing efficiency also attract much attention. This study proposes a replica mechanism based on the Arrangement Graph-based Overlay (AGO), and enhances the joining procedures of the AGO. Enhanced AGO reduces system overhead by reducing large number of polling messages of the joining process. In addition, replica mechanism is also integrated into the enhanced AGO to buttress the efficiency of the searching algorithm. Experimental results of the enhanced AGO have demonstrated that efficient routing as well as less bandwidth consumption during communication can be realized.

CFR: a peer-to-peer collaborative file repository system

Due to the high availability of the Internet, many large cross-organization collaboration projects, such as SourceForge, grid systems etc., have emerged. One of the fundamental requirements of these collaboration efforts is a storage system to store and exchange data. This storage system must be highly scalable and can efficiently aggregate the storage resources contributed by the participating organizations to deliver good performance for users. In this paper, we propose a storage system, Collaborative File Repository (CFR), for large scale collaboration projects. CFR uses peer-to-peer techniques to achieve scalability, efficiency, and ease of management. In CFR, storage nodes contributed by the participating organizations are partitioned according to geographical regions. Files stored in CFR are automatically replicated to all regions. Furthermore, popular files are duplicated to other storage nodes of the same region. By doing so, data transfers between users and storage nodes are confined within their regions and transfer efficiency is enhanced. Experiments show that our replication can achieve high efficiency with a small number of duplicates.

On design and implementation of adaptive data classification scheme for DSM systems

Distributed Shared Memory (DSM) environment is built by using specific softwares, to combine a number of computer hardware resources into one computing environment. Such environment not only provides an easy way to execute parallel applications, but also combines resources to speedup execution of these applications. DSM systems need to maintain data consistency in memory, what usually leads to communication overhead. Therefore, there exist a number of strategies that can be used to overcome this overhead and improve overall performance. Prefetching strategies have been proven to show great performance in DSM systems, since they can reduce data access communication latencies from remote nodes. However, these strategies also transfer unnecessary prefetching pages to remote nodes. In this research paper, we focus on the analysis of data access pattern during execution of parallel applications. We propose an Adaptive Data Classification scheme to improve prefetching strategy, with the goal to improve overall performance. Adaptive Data Classification scheme classifies data according to the access behavior of pages, so that home node uses past history access patterns of remote nodes to decide whether it needs to transfer related pages to remote nodes. From experimental results, our method can improve the performance of prefetching strategies in DSM systems.

An Overlay Network Based on Arrangement Graph with Fault Tolerance

As people change the habit of using the Internet, network technology has become matured. Unlike client-server, peer-to-peer (P2P) technology increases the convenience of people’s daily life. The routing efficiency of P2P system without centralized server always is an important issue. This paper proposes a virtual peer mechanism of P2P overlay network based on the arrangement graph to make exiting physical peers be agent peers for vacant peers. Each vacant peer is managed by a physical peer who often is its neighbor, and the vacant peer is called virtual peer. Physical peers and virtual peers make the arrangement graph full, and make the number of routing hops can be limited within the diameter of the arrangement graph. From experimental results, this system can keep routing efficiency no matter the number of peers and do not increase system overhead.