• Journal of Information Technology Services
• /
• v.7 no.4
• /
• pp.167-178
• /
• 2008

This paper addresses the issue of workflow on Grid and P2P, and proposes a layered workflow architecture and its related workflow models that are used for not only distributing workflows' information onto Grid or P2P resources but also scheduling the enactment of workflows. Especially, the most critical rationale of this paper is on the fact that the nature of Grid computing environment is fitted very well into building a platform for the maximally parallel and very large scale workflows that are frequently found in very large scale enterprises. The layered architecture proposed in this paper, which we call Enterprise Workflow Grid Architecture, is targeting on maximizing the usability of computing facilities in the enterprise as well as the scalability of its underlined workflow management system in coping with massively parallel and very large scale workflow applications.

• Proceedings of the Korea Association of Information Systems Conference
• /
• 2003.11a
• /
• pp.55-63
• /
• 2003

Workflow Management System (WFMS) is a software system to support an efficient execution, control and management of complex business processes. Since traditional commercial systems mainly focus on automating processes, they don't have methods for enhancing the task performer's efficiency. In this paper, we propose a new method of executing business processes more efficiently in that a whole process is scheduled considering the degree of the participants' workload. The method allows managing the largest constraints among constituent resources of the process. We utilize DBR scheduling techniques to develop the method. We first consider the differences between workflow process models and DBR application models, and then develop the modified drum, buffer and rope. This leads us to develop WF-DBR (WorkFlow-DBR) that can control the proper size of the task performers' work list and arrival rate of process instances. Use of WF-DBR improves the efficiency of the whole process as well as the participants' working condition. We then carry out a set of simulation experiments and compare the effectiveness of our approach with that of scheduling techniques used in existing systems.

• ETRI Journal
• /
• v.42 no.3
• /
• pp.388-398
• /
• 2020

We herein propose a heuristic redundancy selection algorithm that combines resubmission, replication, and checkpointing redundancies to reduce the resiliency overhead in fault-tolerant workflow scheduling. The appropriate combination of these redundancies for workflow tasks is obtained in two consecutive phases. First, to compute the replication vector (number of task replicas), we apportion the set of provisioned resources among concurrently executing tasks according to their needs. Subsequently, we obtain the optimal checkpointing interval for each task as a function of the number of replicas and characteristics of tasks and computational environment. We formulate the problem of obtaining the optimal checkpointing interval for replicated tasks in situations where checkpoint files can be exchanged among computational resources. The results of our simulation experiments, on both randomly generated workflow graphs and real-world applications, demonstrated that both the proposed replication vector computation algorithm and the proposed checkpointing scheme reduced the resiliency overhead.

• Asia pacific journal of information systems
• /
• v.11 no.3
• /
• pp.149-164
• /
• 2001

In this paper, an adaptive workflow management system, called K-WFMS, is proposed. The K-WFMS integrates database system and knowledge-based system to automate business processes that are executed with complex and various business rules such as task scheduling, role resolution, and exception handling rules. The K-WFMS is adaptable in the sense that it allows its users to change workflow schema in the course of workflow execution as well as it provides rule-based modeling constructs to handle predictable exceptions during workflow modeling. The overall architecture and implementation of K-WFMS are explained, and the change propagation mechanism to maintain validity of workflow model is suggested.

• The KIPS Transactions:PartA
• /
• v.12A no.4 s.94
• /
• pp.289-296
• /
• 2005

Many jobs in Grid environments consist of several subtasks, and these subtasks can be represented by a workflow, which is executed effectively on a Grid. In this paper, we present Meta services which describe a mapping from a service request to a workflow in Grid environments. By using Meta services, a workflow in Grid environments could adapts various service concepts such as portal services, Grid services, and Web services. And the workflow can be shared and reused among workflow users. Furthermore, historical performance data can be included in Meta services, so effective scheduling of the workflow is also possible.

• Proceedings of the Korea Society of Information Technology Applications Conference
• /
• 2003.06a
• /
• pp.47-48
• /
• 2003

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.8 no.9
• /
• pp.3126-3145
• /
• 2014

Cloud computing is a computing paradigm in which users can rent computing resources from service providers according to their requirements. A spot instance in cloud computing helps a user to obtain resources at a lower cost. However, a crucial weakness of spot instances is that the resources can be unreliable anytime due to the fluctuation of instance prices, resulting in increasing the failure time of users' job. In this paper, we propose a Genetic Algorithm (GA)-based workflow scheduling scheme that can find the optimal task size of each instance in a spot instance-based cloud computing environment without increasing users' budgets. Our scheme reduces total task execution time even if an out-of-bid situation occurs in an instance. The simulation results, based on a before-and-after GA comparison, reveal that our scheme achieves performance improvements in terms of reducing the task execution time on average by 7.06%. Additionally, the cost in our scheme is similar to that when GA is not applied. Therefore, our scheme can achieve better performance than the existing scheme, by optimizing the task size allocated to each available instance throughout the evolutionary process of GA.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.29 no.1
• /
• pp.1-8
• /
• 2006

This paper proposes a real-time resource allocation algorithm for minimizing the completion time of overall workflow process. The jobs in a workflow process are interrelated through the precedence graph including Sequence, AND, OR and Loop control structure. A resource should be allocated for the processing of each job, and the required processing time of the job can be varied by the resource allocation decision. Each resource has several inherent restrictions such as the functional, geographical, positional and other operational characteristics. The algorithm suggested in this paper selects an effective resource for each job by considering the precedence constraint and the resource characteristics such as processing time and the inherent restrictions. To investigate the performance of the proposed algorithm, several numerical tests are performed for four different workflow graphs including standard, parallel and two series-parallel structures. In the tests, the solutions by the proposed algorithm are compared with random and optimal solutions which are obtained by a random selection rule and a full enumeration method respectively.

• 제어로봇시스템학회:학술대회논문집
• /
• 1996.10a
• /
• pp.354-357
• /
• 1996

Virtual Manufacturing System(VMS) is an integrated computer based model which has physical, logical schema and behavior of real manufacturing system. In this paper, an integrated scheduling system is developed to simulate and control a Virtual Factory. A workflow model is constructed to define and analyze the structure of a VMS. On-line dynamic dispatching system is developed using MultiPass algorithm and scheduling system considering dynamic CAPP is carried out. Integrated scheduling system developed in this paper reduces the discrepancies between virtual model and real manufacturing system, and control of real shop floor is possible.

• Journal of Information Processing Systems
• /
• v.8 no.4
• /
• pp.555-566
• /
• 2012

The cloud environment makes it possible to analyze large data sets in a scalable computing infrastructure. In the bioinformatics field, the applications are composed of the complex workflow tasks, which require huge data storage as well as a computing-intensive parallel workload. Many approaches have been introduced in distributed solutions. However, they focus on static resource provisioning with a batch-processing scheme in a local computing farm and data storage. In the case of a large-scale workflow system, it is inevitable and valuable to outsource the entire or a part of their tasks to public clouds for reducing resource costs. The problems, however, occurred at the transfer time for huge dataset as well as there being an unbalanced completion time of different problem sizes. In this paper, we propose an adaptive resource-provisioning scheme that includes run-time data distribution and collection services for hiding the data transfer time. The proposed adaptive resource-provisioning scheme optimizes the allocation ratio of computing elements to the different datasets in order to minimize the total makespan under resource constraints. We conducted the experiments with a well-known sequence alignment algorithm and the results showed that the proposed scheme is efficient for the cloud environment.