Weiqin Tong

A parallel programming environment on grid

Computational grids are environment that provides the ability to exploit diverse, geographically distributed resources. Bulk Synchronous Parallel (BSP) model is a widely used parallel programming model. With rapid development of grid technologies, users need a new environment that can run parallel programs on computational grids. We present here a new implementation of BSP, which is called BSP-G. It constructs a parallel programming environment on computational grids. In our BSP-G environment, users can develop parallel programs with using BSP model and run them on grid. Our BSP-G library uses services provided by the Globus Toolkit for authentication, authorization, resource allocation, executable staging, and I/O, as well as for process creation, monitoring, and control.

Agent Assisted ServiceBSP Model in Grids

The bulk synchronous parallelism (BSP) model is useful in developing scalable software for parallel computation. To adapt the BSP model to the grid environment, the BSP-G, a BSP development library based on the services of Globus Toolkit has been proposed. In this paper, we further the work of BSP-G and introduce an agent assisted ServiceBSP model. The agent in this model could make a QoS-aware selection of computational services and monitor the status of those services. During the selection phase, the agent can handle the diversity of QoS information of services by referring to an ontology repository of quality of service (QoS). An experiment shows that the agent could provide users with satisfied supersteps by selecting Grid services on QoS demand

Agent based ServiceBSP Model with Superstep Service for Grid Computing

This paper proposes an agent based ServiceBSP model with superstep service to support the parallel computing in grid environment. Superstep service is an agent infrastructure handling different types of agents. In this model, the service selection agent generated by superstep service selects computation services with QoS (quality of service) consideration using the IP (integer program) approach. A computing-intensive application is divided into a group of parallel tasks. Each of them is wrapped into a task agent migrating to the local domain of the selected computational services. These task agents are organized in BSP (bulk synchronous parallelism) style, which helps the users to develop performance predictable applications. The experiment shows that the service selection agent can provide the users with satisfied QoS of the applications. And the agents can be generated at low cost.

Toward a Virtual Grid Service of High Availability

A new regulation approach is proposed to obtain a virtual resource service of high availability and service capacity on the basis of resources of low availability and small capacity. Some regulation algorithms with distinct characteristics are introduced.

Component-based integration towards a frequency-regulating home appliance control system

Conventional approaches to embedded software development are very costly, mainly because of their close reliance on application-dependant design, ad-hoc implementation, time-consuming performance tuning and verification. The resulting software is often hard to maintain, upgrade and customize. Component-based integration is an effective method to address the problem. As the result of our effort, an embedded software developing platform is constructed while developing an embedded control system for frequency-regulating home appliances. In this paper, the control system is described, and the developing platform is presented in detail, including an Application Component Library and a Fast Developing Tool. Component-based embedded software design and implementation are also put forward.

Economic Heuristic Guided Price-Regulating Mechanism in SHGRB

The accelerated development in Grid Computing has positioned as promising next generation computing platforms for solving large-scale resource intensive problems. However, the management of resources and scheduling computations in a Grid environment remain complex and immature. The computing economy is argued for the necessity to create a real world scalable Grid because it provides a fair basis in successfully regulating decentralization and heterogeneity presented in Grid environment. Although efforts have been made on the economic mechanisms in Grid, the fixed cost model of resources pricing in current super-scheduler or meta-scheduler and the weakness of load balance in resource scheduling should be improved by any means. Therefore, this paper proposes an economic heuristic guided price-regulating mechanism in the Shanghai Grid Resource Broker (SHGRB) to 1) better adapt to the dynamic changes of grid environment; 2) regulate resource prices for stronger load balance; 3) provide higher quality services for users.

Service organization in OGSI

Virtual organization refers to the flexible, secure, coordinated resource sharing among dynamic collections of individuals, institutions, and resources. In OGSA, resources are represented as grid services. Until now, more efforts are taken to provide specific services for specific VO while ignoring the possible service conflicts or verbosity between different VOs in OGSI, which may cause chaos in the host. To ease the management of services, a distributed multi-layer service organization is proposed in this paper. Meanwhile, we present a new model to support building high-level services. Services collaboration and performance are also considered due to its importance in grid community especially in data grid, current techniques are customized to satisfy the requirements.

Towards a Mobile Service Mechanism in a Grid Environment

With the rapid development of Internet, Grid becomes more and more popular. In this paper, mobile service is introduced to solve some key problems in Grid environment, such as load balance and fault tolerance etc. A mobile service is the combination of Grid service and mobile agent and has migration ability that can make it migrate between grids and ability to detect surroundings to decide when and where to migrate.

Design for Reliable Service Aggregation in an Architectural Environment

How to construct a reliable service aggregation is an important issue in service oriented computing. The paper introduces a design environment to improve the consistence of a service aggregation by cost-effectively using the model checking technique in a software architecture framework.

A Service-Based Hierarchical Architecture for Parallel Computing in Grid Environment

Grid computing technologies enable widespread sharing and coordinated us of networked resources, including computing resources. Open Grid Service Architecture (OGSA) marries Web services to grid protocols, thereby making progress in defining interfaces for grid services. We put forward here a hierarchical architecture, which is based on grid service, for parallel computing in grid environment. Several grid services are proposed in this paper for building our architecture, including cooperation service, resource information service, computing service, communication service, and other atomic services such as MOM service. The technology of building high layer grid service is also introduced in this paper.