Feng-Jian Wang

Online scheduling of workflow applications in grid environments

Scheduling workflow applications in grid environments is a great challenge, because it is an NP-complete problem. Many heuristic methods have been presented in the literature and most of them deal with a single workflow application at a time. In recent years, several heuristic methods have been proposed to deal with concurrent workflows or online workflows, but they do not work with workflows composed of data-parallel tasks. In this paper, we present an online scheduling approach for multiple mixed-parallel workflows in grid environments. The proposed approach was evaluated with a series of simulation experiments and the results show that the proposed approach delivers good performance and outperforms other methods under various workloads.

A distributed server architecture supporting dynamic resource provisioning for BPM-oriented workflow management systems

Workflow management systems have been widely used in many business process management (BPM) applications. There are also a lot of companies offering commercial software solutions for BPM. However, most of them adopt a simple client/server architecture with one single centralized workflow-management server only. As the number of incoming workflow requests increases, the single workflow-management server might become the performance bottleneck, leading to unacceptable response time. Development of parallel servers might be a possible solution. However, a parallel server architecture with a fixed-number of servers cannot efficiently utilize computing resources under time-varying system workloads. This paper presents a distributed workflow-management server architecture which adopts dynamic resource provisioning mechanisms to deal with the probable performance bottleneck. We implemented a prototype system of the proposed architecture based on a commercial workflow management system, Agentflow. A series of experiments were conducted on the prototype system for performance evaluation. The experimental results indicate that the proposed architecture can deliver scalable performance and effectively maintain stable request response time under a wide range of incoming workflow request workloads.

An incremental analysis for resource conflicts to workflow specifications

Workflow management technology helps modulizing and controlling complex business processes within an enterprise. Generally speaking, a workflow management system (WfMS) is composed of two primary components, a design environment and a run-time system. Structural, timing and resource verifications of a workflow specification are required to assure the correctness of the specified system. In this paper, an incremental methodology is constructed to analyze resource consistency and temporal constraints after each edit unit defined on a workflow specification. The methodology introduces several algorithms for general and temporal analyses. The output returned right away can improve the judgment and thus the speed and quality on designing.

A process-centered software engineering environment with network centric computing

A good process-centered software engineering environment, PSEE, helps the project teams develop software. To support concurrent teamwork, a distributed PSEE supports the interaction and cooperation of project members. The Internet and Web technologies have brought up a new concept called network centric computing, NCC. The NCC provides a standard interface for accessing data to improve the extensibility of applications. We present the design and implementation of a distributed PSEE with NCC. The architecture of our PSEE is divided into three layers: kernel, intermediate and user interface layers. These layers contain the web server, process server and object server which communicate with each other through a stateless access protocol, SAP. These servers are in charge of process modeling and execution, while the web browser is in charge of the user interface such as an agenda. We present the activities of process enactment within a project. We make a comparison between our PSEE with the NCC model and prior PSEE with the client-server model.

Dynamic resource provisioning for interactive workflow applications on cloud computing platform

Cloud computing opens new opportunities for application providers because with the policy add as needed and pay as used they can economize the cost for computing resources. In cloud environments, issues such as resource allocation and dynamic resource provisioning based on users QoS constraints are yet to be addressed for interactive workflow applications. This paper develops an effective load metric, remaining tasks, for interactive workflow applications. Based on this metric load dispatching and dynamic resource provisioning approaches are proposed which outperform existing methods under a series of simulation evaluations. Experimental results show that the proposed approaches offer application providers better maintenance of QoS-satisfied response time under time-varying workload, at the minimum cost of resource usage.

Detecting artifact anomalies in business process specifications with a formal model

Many business process analysis models have been proposed, however there are few discussions for artifact usages in workflow specifications. A well-structured business process with sufficient resources might fail or yield unexpected results dynamically due to inaccurate artifact specification, e.g. an inconsistency between artifact and control flow, or contradictions between artifact operations. This paper, based on our previous work, presents a model for describing the input/output of a workflow process and analyzes the artifact usages upon the model. This work identifies and formulates thirteen cases of artifact usage anomalies affecting process execution and categorizes the cases into three types. Moreover, the methods for detecting these anomalies with time complexities O(n^2), less than O(n^3) in previous methods, are presented. Besides, the paper uses an example to demonstrate the processing of them.

A Web database application model for software maintenance

Web database application programs may consist of hundreds of files. The interoperations among these files could be complex, and most of them are lack of design documents. When a programmer maintains (changes) some programs of an application, the program files affected by the changes need be updated simultaneously to keep the functional consistency. One major difficulty of maintenance is how to identify these affected parts, especially in Web database applications. In this paper, we present a model, called HED model, for the maintenance of Web database applications. HED model decomposes a Web database application into three diagrams, hyperlink diagrams, entity-relationship diagrams, and data-flow diagrams, which are used to represent different aspects of a Web database application. Based on the HED model, the programs files affected by a program change can be identified precisely via the structure and database analyses. In addition, a maintenance tool is implemented to demonstrate the capability of the HED model.

Scheduling online mixed-parallel workflows of rigid tasks in heterogeneous multi-cluster environments

Workflow scheduling on parallel systems has long been known to be a NP-complete problem. As modern grid and cloud computing platforms emerge, it becomes indispensable to schedule mixed-parallel workflows in an online manner in a speed-heterogeneous multi-cluster environment. However, most existing scheduling algorithms were not developed for online mixed-parallel workflows of rigid data-parallel tasks and multi-cluster environments, therefore they cannot handle the problem efficiently. In this paper, we propose a scheduling framework, named Mixed-Parallel Online Workflow Scheduling (MOWS), which divides the entire scheduling process into four phases: task prioritizing, waiting queue scheduling, task rearrangement, and task allocation. Based on this framework, we developed four new methods: shortest-workflow-first, priority-based backfilling, preemptive task execution and All-EFT task allocation, for scheduling online mixed-parallel workflows of rigid tasks in speed-heterogeneous multi-cluster environments. To evaluate the proposed scheduling methods, we conducted a series of simulation studies and made comparisons with previously proposed approaches in the literature. The experimental results indicate that each of the four proposed methods outperforms existing approaches significantly and all these approaches in MOWS together can achieve more than 20% performance improvement in terms of average turnaround time. A scheduling framework for mixed-parallel workflows of rigid tasks is proposed.The framework highlights four scheduling phases for further research directions.Four task scheduling methods are developed for optimizing execution performance.

Design patterns for parallel computations of master-slave model

This paper presents two design patterns useful for parallel computations of master-slave model. These patterns are concerned with task management and parallel and distributed data structures. They can be used to help address the issues of data partition and mapping, dynamic task allocation, and load balancing in parallel programming with the benefit of less programming efforts and better program structures. The patterns are described in object-oriented notation, accompanied with illustrative examples in C++. We also provide our experience in applying these patterns to two scientific simulation programs simulating Ising model and plasma respectively. Since master-slave model is a widely used parallel programming paradigm, the design patterns presented in this paper have large potential application in parallel computations.

Developing high-performance scientific applications in distributed computing environments

This paper presents our experience and future plan for exploring high-performance scientific computing in distributed computing environments. The purpose of our research is to ease the development of high-performance scientific applications in distributed computing environments. This paper addresses issues arising when applying object-oriented programming techniques to distributed scientific applications.