Pill-Woo Lee

Reconfiguration Mechanisms for Virtual Organization Using Remote Deployment of Grid Services

Grid computing is becoming more and more attractive for coordinating large-scale heterogeneous resource sharing and problem solving. Although Globus toolkit version 4 which is developed based on WSRF can successfully deal with the heterogeneity of resources and the diversity of applications, the current grid environments based on Globus toolkit version 4 have several limitations regarding dynamic reconfiguration of virtual organization (VO) and remote hot deployment of grid services. This paper proposes WIVO (Web-based user Interface for Virtual Organization), the web-based user interface using web service. Using WIVO, user requested sub-VO can easily be configured on top of grid environment based on Globus toolkit version 4, and grid services can also be deployed remotely for better utilization of resources and applications. It is expected that the VO reconfiguration function as well as the remote grid service deployment function provided in WIVO can improve resource and application utilization.

Cyber Physical System-Based Convergence Operation of Data Intensive Computing Resources

Large-scale scientific experiments involve a large number of computers. These computers generally employ grid computing to perform non-interactive, data-intensive tasks. Recent scientific studies have required multiple computers to handle massive data sets. An efficient method of supporting various research is to provide computing resources through a unified data center. This has complicated the management of computing resources in the data center. The deployment and the status of computing resources change frequently according to the requirements of each research, renewal of research support contracts, and adjustment of computing resources. This study presents a cyber physical system based on convergence operation of data-intensive computing resources to support various research groups. The proposed operation takes into account the modification of computing resources, other changes resulting from such modification, and information reflecting modification of the physical world. We first present the workload of operations and related information. Based on these relationships, we derive an optimized work process and the required physical information. For computing resources including existing equipment, we use a mobile device and a wireless network to facilitate conversion between the physical world and cyber world. The mobile device of administrator captures sequence of remote controlled management indicator lights to detect physical modification.

An efficient information dissemination for publish/subscription system on grid

In ubiquitous environment, with rapid advances being made in sensor technology, wireless sensor networks (WSNs) have emerged to collect information from real world. For sharing huge amount of sensor data efficiently with divers users, an information dissemination mechanism is very necessary and important component. In this paper, we have proposed the information dissemination architecture, which is implemented using OGSA-DAI and web services based on Info-D standard, and an efficient matching algorithm, CGIM, for u-health care system. The information dissemination system supports scheduling for both time and event driven delivery and EAQ for 1:1 event and n:1 event handling. From CGIM algorithm, matching time is significantly reduced for information dissemination, and the algorithm allows multiple events with ranged predicates, flexible expression for diverse consumers, and efficient matching and partial optimization on pub/sub system.

An optimal and dynamic monitoring interval for grid resource information system

Grid technology uses geographically distributed resources from multiple domains. For that reason, resource monitoring services or tools will run on various kinds of systems to find static resource information and dynamic resource information, such as architecture vendor, OS name and version, MIPS rate, memory size, CPU capacity, disk size, NIC information, CPU usage, network usage (bandwidth, latency), and memory usage, etcs. Thus monitoring itself may cause the system overhead. This paper proposes the optimal monitoring interval to reduce the cost of monitoring services and the dynamic monitoring interval to measure the monitoring events accurately. By simulating and implementing those two factors, we find out that unnecessary system overhead is significantly reduced and accuracy of events is satisfied.

EIMD: a new congestion control for fast long-distance networks

In fast long-distance networks, TCP fails to fully utilize the bandwidth due to its congestion control mechanism A plethora of congestion control schemes that may enhance the performance of the transport protocol in fast long-distance networks have been proposed The proposed schemes aim to satisfy three requirements of congestion control schemes: bandwidth scalability, TCP friendliness, and RTT fairness However, due to the trade-off among these requirements, it is difficult to satisfy all the requirements simultaneously In this paper, we propose a new window-based congestion control scheme called EIMD (Exponential Increase/Multiplicative Decrease) that increases congestion window size exponentially to quickly grap the unutilized bandwidth We evaluate the performance of EIMD via computer simulations The simulation results show that EIMD satisfies the three requirements In addition, EIMD converges fast to the fair-share points.

The Design and Implementation of an Available Bandwidth Measurement Scheme in the K * Grid System

Grid computing is an emerging technology that enables global resource sharing. In Korea, the K*Grid provides an extremely powerful research environment to both industries and academia. As part of the K*Grid project, we have constructed, together with the Korea Institute of Science and Technology Information and a number of domestic universities, a supercomputer Grid test bed which connects several types of supercomputers based on the globus toolkit. To achieve efficient networking in this Grid testbed, we propose a novel method of available bandwidth measurement, called Decoupled Capacity measurement with Initial Gap (DCIG), using packet trains. DCIG can improve the network efficiency by selecting the best path among several candidates. Simulation results show that DCIG outperforms previous work in terms of accuracy and the required measurement time. We also define a new XML schema for DCIG request/response based on the schema defined by the Global Grid Forum (GGF) Network Measurement Working Group (NM-WG).

OGSA-GFS : A OGSA based Grid File System

Since many grid applications access data from file system or database systems, the management of data within grids is an essential problem. In this work, we address the necessity of grid file system and propose OGSA based grid file system(OGSA-GFS). OGSA-GFS integrates and manages distributed files and replicated files across multiple institutions in Grids. And it provides the global view of the files to users. This global view is used to access the distributed files for grid user or grid application regardless of the physical filename and host address. In order to provide the global view and standard interface, we design three main services, the GFS namespace service, the GFS replica management service, and the GFS information service, and implement these services on the OGSA..