Execution coordination in mobile agent-based distributed job workflow execution

Mobile agent-based distributed job workflow execution requires the use of execution coordination techniques to ensure that an agent executing a subjob can locate its predecessors’ execution results. This paper describes the classification, implementation, and evaluation of execution coordination techniques in the mobile agent-based distributed job workflow execution system. First, a classification of the existing execution coordination techniques is developed for mobile agent systems. Second, to put the discussion into perspective, our framework for mobile agent-based distributed job workflow execution over the Grid (that is, MCCF: Mobile Code Collaboration Framework) is described. How the existing coordination techniques can be applied in the MCCF is also discussed. Finally, a performance study has been conducted to evaluate three coordination techniques using real and simulated job workflows. The results are presented and discussed in the paper.     

Mobile agent‐based computational steering for distributed applications

The mobile agent‐based computational steering (MACS) for distributed applications is presented in this article. In the MACS, a mobile agent platform, Mobile‐C, is embedded in a program through the Mobile‐C library to support C/C++ mobile agent code. Runtime replaceable algorithms of a program are represented as agent services in C/C++ source code and can be replaced with new ones through mobile agents. In the MACS, a mobile agent created and deployed by a user from the steering host migrates to computing hosts successively to replace algorithms of running programs that constitute a distributed application without the need of stopping the execution and recompiling the programs. The methodology of dynamic algorithm alteration in the MACS is described in detail with an example of matrix operation. The Mobile‐C library enables the integration of Mobile‐C into any C/C++ programs to carry out computational steering through mobile agents. The source code level execution of mobile agent code facilitates handling issues such as portability and secure execution of mobile agent code. In the MACS, the network load between the steering and computing hosts can be reduced, and the successive operations of a mobile agent on multiple computing hosts are not affected whether the steering host stays online or not. The employment of the middle‐level language C/C++ enables the MACS to accommodate the diversity of scientific and engineering fields to allow for runtime interaction and steering of distributed applications to match the dynamic requirements imposed by the user or the execution environment. An experiment is used to validate the feasibility of the MACS in real‐world mobile robot applications. The experiment replaces a mobile robot's behavioral algorithm with a mobile agent at runtime. Copyright © 2009 John Wiley & Sons, Ltd.     

A Direct Execution Approach to Simulating Mobile Agent Algorithms

Mobile agent technology has been applied to develop the solutions for various kinds of parallel and distributed computing problems. However, performance evaluation of mobile agent algorithms remains a difficult task, mainly due to the characteristics of mobile agents such as distributed and asynchronous execution, autonomy and mobility. This paper proposes a general approach based on direct execution simulation for evaluating the performance of mobile agent algorithms by collecting and analyzing the information about the agents during their execution. We describe the proposed generic simulation model, named MADES, the architecture of a software environment based on MADES, and a prototype implementation. A mobile agent-based distributed load balancing algorithm has been used for experiments with the prototype.     

基于移动代理的工作流原型研究与设计

快速发展的移动代理和网格技术给工作流提供了新的集成各种分布式资源的计算环境。传统的工作流架构有很多弱点,比如在动态环境中缺少灵活性等。提出了一个集成Web Services,网格,工作流和移动代理技术的分布式工作流架构,可以通过移动代理间的通讯来达到工作流之间协同工作,而且这个架构增强了工作流的灵活性和可靠性,以这个新的模型为基础实现了一个具体的工作流工作站,且可实现系统的安全使用。     

Using mobile agents to support interorganizational workflow management

This article argues that the mobile agent approach is well suited for sporadic communication in open distributed systems, especially for rather ''loose'' cooperations across local and organizational boundaries. In an increasing number of cases, management of distributed business procedures reaches beyond such boundaries. This means, for most existing workflow management systems, that cooperating partners are required to give up their local autonomy. However, for cases in which business partners intend to cooperate but still need to preserve their local autonomy, process participation on the basis of mobile agents represents an appropriate mechanism. This article shows how such process integration can be achieved. It also demonstrates how the Common Open Service Market (COSM) system software can be extended in order to use petri-net based process definitions that realize mobile agents in an integrated distributed system platform.     

Job scheduling and processor allocation for grid computing on metacomputers

Scheduling is a fundamental issue in achieving high performance on metacomputers and computational grids. For the first time, the job scheduling problem for grid computing on metacomputers is studied as a combinatorial optimization problem. A cost model is proposed for modeling communication heterogeneity on computational grids. A processor allocation algorithm is developed which always finds an optimal processor allocation that minimizes the effective execution time of a job when the job is being scheduled. It is proven that the list scheduling (LS) algorithm can achieve reasonable worst-case performance bound in grid environments supporting distributed supercomputing with large applications. We compare the performance of various job scheduling and processor allocation algorithms for grid computing on metacomputers. We evaluate the performance of 128 combinations of two job scheduling algorithms, four initial job ordering strategies, four processor allocation algorithms, and four metacomputers by extensive simulation. It is found that the combination of largest job first (LJF) initial job ordering and minimum effective execution time (MEET) or largest machine first (LMF) processor allocation algorithm yields the best average-case performance, and the choice of FCFS and LS depends on the range of job sizes. It is also observed that communication heterogeneity does have significant impact on schedule lengths.     

分布式工作流执行服务的设计与实现

工作流执行服务的实现方案是实施工作流技术最为关键的一个方面,工作流管理系统的应用特点决定了它必须与各种分布的、异构的应用进行通信或与其集成。为了给用户提供透明的工作流执行服务而隐藏工作流执行服务实现的复杂性,文章提出了一种具有三层层次结构的工作流执行服务实现方案,其中的分布式通信服务层由通信中间件实现,使开发者无需关心程序间通信的复杂性,从而将精力放在解决业务问题和优化业务逻辑方面,提高了应用开发的效率。     

An agent-oriented approach to process partition and planning in migrating workflow systems

Mobile agents act as the task executors in migrating workflow system. As the size of workflow is increased by including many tasks and branches, multiple mobile agents should be used where each agent is responsible for a branch of the workflow process and fulfills the workflow goal by cooperating with its partners. In this case, the workflow process needs to be partitioned into a set of sub-processes before execution so that each sub-process can be assigned to one mobile agent. This paper proposes a structured process partition approach that includes process structure partition and Quality of Service(QoS) objective partition. The former partitions a structured process into a set of sub-processes with dominant relations and each sub-process consists of a sequence of tasks. The latter distributes QoS objectives, such as the expected budget or deadline of the whole workflow, over all of the sub-processes. In addition, a sub-process execution planning algorithm with QoS objective based on Markov Decision Process (MDP) is also put forward in this paper. The experiment results show that the effects of workflow partition method and MDP based process planning method are sound.     

云中截止时间动态分配的工作流调度成本优化算法

现如今,如何在满足截止时间约束的前提下降低工作流的执行成本,是云中工作流调度的主要问题之一。三步列表调度算法可以有效解决这一问题。但该算法在截止时间分配阶段只能形成静态的子截止时间。为方便用户部署工作流任务,云服务商为用户提供了的三种实例类型,其中竞价实例具有非常大的价格优势。为解决上述问题,提出了截止时间动态分配的工作流调度成本优化算法(S-DTDA)。该算法利用粒子群算法对截止时间进行动态分配,弥补了三步列表调度算法的缺陷。在虚拟机选择阶段,该算法在候选资源中增加了竞价实例,大大降低了执行成本。实验结果表明,相较于其他经典算法,该算法在实验成功率和执行成本上具有明显优势。综上所述,S-DTDA算法可以有效解决工作流调度中截止时间约束的成本优化问题。     

网格环境中基于移动Agent的工作流架构研究

传统的工作流架构有很多弱点,比如架构严重降低了系统性能,在动态环境中缺少灵活性。本文提出了一个集成了WebServices,网格,工作流和移动代理技术的分布式工作流架构。在这个架构中,我们可以通过移动代理间的通讯来达到工作流之间协同工作,而且这个架构增强了工作流的灵活性和可靠性。最后,以这个新的模型为基础实现了一个具体的工作流工作站。     

AgentTeamwork: Coordinating grid-computing jobs with mobile agents

AgentTeamwork is a grid-computing middleware system that dispatches a collection of mobile agents to coordinate a user job over remote computing nodes in a decentralized manner. Its utmost focus is to maintain high availability and dynamic balancing of distributed computing resources to a parallel-computing job. For this purpose, a mobile agent is assigned to each process engaged in the same job, monitors its execution at a different machine, takes its periodical execution snapshot, moves it to a lighter-loaded machine, and resumes it from the latest snapshot upon an accidental crash. The system also restores broken inter-process communication involved in the same job using its error-recoverable socket and mpiJava libraries in collaboration among mobile agents. We have implemented the first version of our middleware including a mobile agent execution platform, error-recoverable socket and mpiJava API libraries, a job wrapper program, and several types of mobile agents such as commander, resource, sentinel, and bookkeeper agents, each orchestrating, allocating resources to, monitoring and maintaining snapshots of a user process respectively. This paper presents AgentTeamwork’s execution model, its implementation techniques, and our performance evaluation using the Java Grande benchmark test programs. Munehiro Fukuda received a B.S. from the College of Information Sciences and an M.S. from the Master’s Program in Science and Enginnering at the University of Tsukuba in 1986 and 1988. He received his M.S. and Ph.D. in Information and Computer Science at the University of California at Irvine in 1995 and 1997, respectively. He worked at IBM Tokyo Research Laboratory from 1988 to 1993 and taught at the University of Tsukuba from 1998 to 2001. Since 2001, he has been an assistant professor at Computing & Software Systems, the University of Washington, Bothell. His research interests include mobile agents, multi-threading, cluster computing, grid computing and distributed simulations. Koichi Kashiwagi received a Bachelor of Science degree from the Faculty of Science, Ehime University in 2000 and a Master of Engineering degree from the Department of Compter Science, Ehime University in 2002. In 2004 he became a research assistant in Department of Compter Science, Ehime University. His research interests include distributed computing, job scheduling, and grid computing. Shin-ya Kobayashi received the B.E. degree, M.E. degree, and Dr.E. degree in Communication Engineering from Osaka University in 1985, 1988, and 1991 respectively. From 1991 to 1999, he was on the faculty of Engineering at Kanazawa University, Japan. From 1999 to 2004, He was an Associate Professor in the Department of Computer Science, Ehime University. He is a Professor at Graduate School of Science and Engineering, Ehime University. His research interests include distributed processing, and parallel processing. He is a member of the Information Processing Society of Japan, the Institute of Electrical Engineers of Japan, IEEE, and ACM.     

Algorithm visualization for distributed environments

This paper investigates the visualization of distributed algorithms. We present a conceptual model and a system, VADE, that realizes this model. Since in asynchronous distributed systems there is no way of knowing (let alone, visualizing) the “real” execution, we show how to generate a visualization which is consistent with the execution of the distributed algorithm. We also present the design and implementation of our system. VADE is designed so that the algorithm runs on the server's machines while the visualization is executed on a web page on the client's machine. Programmers can write animations quickly and easily with the assistance of VADE's libraries. End-users can experiment with the animation and change its visual properties.     

A security protocol for mobile agents based upon the cooperation of sedentary agents

Despite its many benefits, mobile agent technology results in significant security threats from agents and hosts. This paper presents a protocol which protects mobile agents from malicious hosts. This protocol combines four concepts: the cooperation between a mobile agent and a sedentary agent; the reference execution (reliable platforms which shelter our cooperating sedentary agents); the cryptography and the digital signature to ensure safe inter-agent communication and time-limited execution (timeout). A dynamic approach which makes use of a timer to make it possible to detect a mobile agent's code re-execution was used. The attack on agent permanent modification was also dealt with. Moreover, the protocol is sufficiently robust so that it is durable and fault tolerant.     

基于移动Agent的个性化数字学习模型研究

在开放、动态的学习资源网络环境中,个性化服务支持对在线学习者尤其重要.讲述了在基于web服务技术的分布式学习环境中如何构造一个分布式的个性化数宁学习环境.在建立服务代理的功能模型的基础上,构造了一种基于多移动Agent的个性化数字学习(E-Learning)框架模型,阐述了其工作流程并进行了结构分析;引入Petri网模型,设计了基于移动Agent的联邦组建与动态服务合成算法;说明了系统实现的技术手段与方法.从对系统模型雏形的应用与性能监测来看,系统模型实现切实可行且运行性能良好.     

Competitive analysis of caching in distributed databases

This paper makes two contributions. First, we introduce a model for evaluating the performance of data allocation and replication algorithms in distributed databases. The model is comprehensive in the sense that it accounts for I/O cost, for communication cost, and, because of reliability considerations, for limits on the minimum number of copies of the object. The model captures existing replica-management algorithms, such as read-one-write-all, quorum-consensus, etc. These algorithms are static in the sense that, in the absence of failures, the copies of each object are allocated to a fixed set of processors. In modern distributed databases, particularly in mobile computing environments, processors will dynamically store objects in their local database and will relinquish them. Therefore, as a second contribution of this paper, we introduce an algorithm for automatic dynamic allocation of replicas to processors. Then, using the new model, we compare the performance of the traditional read-one-write-all static allocation algorithm to the performance of the dynamic allocation algorithm. As a result, we obtain the relationship between the communication cost and I/O cost for which static allocation is superior to dynamic allocation, and the relationships for which dynamic allocation is superior     

Development of a class of distributed termination detectionalgorithms

Accessing and updating information in a self organizing data structure in a distributed environment requires execution of various distributed algorithms. Design of such algorithms is often facilitated by the use of a distributed termination detection algorithm superimposed on top of another distributed algorithm. The problem of distributed termination detection is considered, and message counting is introduced as an effective technique in designing such algorithms. A class of efficient algorithms, based on the idea of message counting, for this problem is presented. After termination has occurred, it is detected within a small number of message communications. These algorithms do not require the FIFO (first in, first out) property for the communication lines. Assumptions regarding the connectivity of the processes are simple. The algorithms are incrementally developed, i.e. a succession of algorithms leading to the final algorithms is presented     

Agent scheduling model for adaptive dynamic load balancing in agent-based distributed simulations

Agent-based distributed simulations are confronted with load imbalance problem, which significantly affects simulation performance. Dynamic load balancing can be effective in decreasing simulation execution time and improving simulation performance. The characteristics of multi-agent systems and time synchronization mechanisms make the traditional dynamic load balancing approaches not suitable for dynamic load balancing in agent-based distributed simulations. In this paper, an adaptive dynamic load balancing model in agent-based distributed simulations is proposed. Due to the complexity and huge time consuming for solving the model, a distributed approximate optimized scheduling algorithm with partial information (DAOSAPI) is proposed. It integrates the distributed mode, approximate optimization and agent set scheduling approach. Finally, experiments are conducted to verify the efficiency of the proposed algorithm and the simulation performance under dynamic agent scheduling. The experiments indicate that DAOSPI has the advantage of short execution time in large-scale agent scheduling, and the distributed simulation performance under this dynamic agent scheduling outperforms that under static random agent distribution.     

一个调度Fork-Join任务图的新算法

任务调度是影响工作站网络效率的关键因素之一.Fork-Join任务图可以代表很多并行结构,但其他已有调度Fork-Join任务图算法忽略了在非全互连工作站网络环境中通信之间不能并行执行的问题,有些效率高的算法又没有考虑节省处理器个数的问题.因此,专门针对该任务图,综合考虑调度长度、非并行通信和节省处理器个数问题,提出了一个基于任务复制的静态调度算法TSA_FJ.通过随机产生任务的执行时间和通信时间,生成了多个Fork-Join任务图,并且采用TSA_FJ算法和其他调度算法对生成的任务图进行调度.结果表明,     

Dependable Grid Workflow Scheduling Based on Resource Availability

Due to the highly dynamic feature, dependable workflow scheduling is critical in the Grid environment. Various scheduling algorithms have been proposed, but seldom consider the resource reliability. Current Grid systems mainly exploit fault tolerance mechanism to guarantee the dependable workflow execution, which, however, wastes system resources. The paper proposes a dependable Grid workflow scheduling system (called DGWS). It introduces a Markov Chain-based resource availability prediction model. Based on the model, a reliability cost driven workflow scheduling algorithm is presented. The performance evaluation results, including the simulation on both parametric randomly generated DAGs and two real scientific workflow applications, demonstrate that compared to present workflow scheduling algorithms, DGWS improves the success ratio of tasks and diminishes the makespan of workflow, so improves the dependability of workflow execution in the dynamic Grid environments.     

A bio-inspired mobile agent-based integrated system for flexible autonomic job shop scheduling

This paper presents a bio-inspired mobile agent-based integrated system for flexible autonomic job shop scheduling. The system matches the autonomic system architecture, inspired by the autonomic nervous system and proposed by the IBM, and has the IBM-defined fundamental self-managing properties, so that it can manage itself with little human intervention. The system conforms to the IEEE FIPA (Foundation for Intelligent Physical Agents) standard. Therefore, the interoperability between agents of the system and agents from many active heterogeneous FIPA compliant agent platforms can be ensured. The system supports the execution of C/C++ mobile agent codes. Thus, it is applicable to a variety of applications, especially for distributed mechatronic and embedded systems. In addition, since the system is composed of agents, including stationary and mobile agents, the system has a high scalability and flexibility to integrate and adopt various scheduling models and algorithms for different scheduling requirements. An overall architecture of the system and critical implementation details about the agency and agents in the system are presented in this article. An energy saving job shop scheduling example is used to validate one autonomic property of the system.