Terence Hung

An integrated and adaptive decision-support framework for high-tech manufacturing and service networks

This article gives an update on a major international collaborative project under the integrated manufacturing and service systems initiative pursued by the Agency for Science, Technology and Research in Singapore. The objective of the project is to investigate how design, analysis, enhancement and implementation of critical business processes in a manufacturing and service network can be realized using one single simulation/application framework. The overall architecture of the framework outlines how commercial simulation packages and web service-based business process application components can be connected through a commercial application framework to achieve maximum leverage and re-usability of the application components involved. The project also addresses research issues with regard to mechanisms for interoperation between commercial simulation packages, symbiotic interaction between simulation-based decision support components and physical systems, and simulation speed-up through multi-objective optimal computing budget allocation techniques on a grid/cluster infrastructure.

The design and implementation of an OGSA-based grid information service

The information service is a key component of a grid environment and critical to the operation of a computational grid. In this work, an OGSA (Open Grid Services Architecture) based information service that complies with OGSI (Open Grid Services Infrastructure) is presented. The main functionality of this information service is the provision of information essential for applications running on a computational grid such as resource information, job status, resource workload, service meta-information, and queue status. This OGSI-compliant information service is built on Globus Toolkit MDS-3, and it works with meta-scheduling services and local job scheduling systems to support resource discovery, job scheduling, and execution management. In this paper, the architecture of the Information Service and the models of information data organization are presented. Some implementation issues are discussed as well.

Design of an SLA-Driven QoS Management Platform for Provisioning Multimedia Personalized Services

In this paper, we present an SLA-driven QoS Management Platform which establishes service level agreement between the service provider and the customers before the service delivery and manages the distributed resources efficiently to support QoS-aware multimedia personalized services (MPS) over large scale networks. Based on service oriented architecture, such a QoS management platform consists of SLA negotiation, QoS management and resource management. The QoS management connects the SLA negotiation and resource management to real-time monitor QoS and adapt the quality levels to cope with the varying network conditions. Such a QoS management platform allows clients to specify the QoS requirements, and enables service providers to offer different QoS levels, negotiate with their customers on the possible quality levels and adapt the resources allocation to optimize the system overall performance. We constructed a prototype system and study the QoS management in both static and dynamic service modes. Experiments are conducted to examine the system communication overhead caused by the SLA communication, and system performance by integrating various strategies for QoS management. The results show that our proposed SLA- driven QoS management platform is highly flexible, extensible and efficient to support QoS-aware MPS over large scale networks.

Architecture Model for Information Service in Large Scale Grid Environments

The Information Service is a core component in the Grid software infrastructure. It provides diverse information to users or other service components in Grid environments. In this paper, we propose an Information Service architecture model for information management in a Grid Virtual Organization (VO). This Information Service is a hierarchical structure which consists of the VO layer, site layer and resource layer: at the resource layer, information agents and pluggable information sensors are deployed on each resource monitored. This information agent and sensor approach provides a flexible framework that enables specific information to be captured; at the site layer, a site information service component with caching capability aggregates and maintains up-to-date information of all the resources monitored within an administrative domain; at the VO layer, a peer-to-peer approach is used to build a virtual network of site information services for information discovery and query in a large scale Grid VO. This decentralized approach makes information management scalable and robust. Our Information Service has been implemented based on the Globus Toolkit 4 as a Web service compliant to the Web Services Resource Framework (WSRF) specifications. The experimental results show that the Information Service presents satisfactory scalability in handling information for large scale Grids.

An OGSI-compliant grid information service - its architecture and performance study

The information service is a key component of a grid environment and critical to the operation of a computational grid. This work presents an information service that complies with the OGSI (open grid services infrastructure) framework. This information service is a hierarchical structure which consists of the virtual organization level, site level and node level. We propose two models of information data organization for the information service, i.e. source-data based model and meta-data based model. Two sets of experiments were conducted to evaluate the performance of the information service. The experimental results show that the amount of data communicated between clients and the information service, and among the component services is the key factor to influence the scalability of the information service. We also found that the meta-data based model is an effective solution to improve the scalability of the information service.

Performance modeling for runtime kernel adaptation: A case study on infectious disease simulation

In many large-scale scientific applications, there may be a compute intensive kernel that largely determines the overall performance of the application. Sometimes algorithmic variations of the kernel may be available and a performance benefit can then be gained by choosing the optimal kernel at runtime. However, it is sometimes difficult to choose the most efficient kernel as the kernel algorithms have varying performance under different execution conditions. This paper shows how to construct a set of performance models to explore and analyze the bottleneck of an application. Furthermore, based on the performance models, a theoretical method is proposed to guide the kernel adaptation at runtime. A component-based large-scale infectious disease simulation is used to illustrate the method. The performance models of the different kernels are validated by a range of experiments. The use of runtime kernel adaptation shows a significant performance gain.

A Multi-Agent Method for Streaming Quality Monitoring and Analysis over Media Grid

Media streaming is a promising technology which enables fast and adaptive delivery of media contents and allows clients to view videos online without waiting for the entire multimedia files to be completely downloaded. In this paper, we address the issue of streaming quality measurement and assessment over a large scale network. We propose a multiagent system for streaming quality monitoring and analysis (MASQMA) which can real-time monitor and analyze the quality of media streaming over a large scale IP network. Based on a decentralized infrastructure, such a system is scalable and flexible to meet versatile streaming applications without increasing the processing burden on the current media streaming systems. It allows users to real-time monitor the QoS with a variety of quality metrics. We conduct experiments and study the performance under various situations (e.g., bandwidth fluctuations, packet loss, and frame jitter) which are frequently encountered in Wide Area Networks. The results show the multi-agent architecture is practical and flexible to monitor and analyze streaming quality in real time.

Information management for computational grids

Information Service is a key component of a Grid environment and crucial to the operation of a computational Grid. This paper presents an Information Service for a computational Grid Virtual Organization (VO). The main functionality of this Information Service is the provision of information essential for applications running on a computational Grid. This Information Service is a hierarchical structure that consists of VO layer, site layer, and resource layer. We propose different models of information data organization for Information Service, and experiments were conducted to evaluate the performance of these models. Based on the experimental results, we further introduce the data organization model for our Information Service. The implementation of the Information Service is based on the Globus Toolkit and complies with the OGSI (Open Grid Services Infrastructure) specifications. In this paper, some application experiences of using the Information Service are also presented.