An Active Resource Management System for Computational Grid*

Size: px
Start display at page:

Download "An Active Resource Management System for Computational Grid*"

Transcription

1 An Active Resource Management System for Computational Grid* Xiaolin Chen 1, Chang Yang 1, Sanglu Lu 2, and Guihai Chen 2 1 Department of Computer Science, Chuxiong Normal University, Chuxiong , China {chenxl,yc}@cxtc.edu.cn 2 State Key Laboratory of Novel Software Tech., Nanjing University, Nanjing , China {sanglu,gchen}@nju.edu.cn Abstract. In this paper, we propose an active grid resource management system supported by active networks for computational grid. First, we construct a scalable two-level resource management architecture. In this architecture, resources in the system are divided into multiple autonomous domains. In each domain, an active resource tree (ART) is organized with resources as its leaf nodes and active routers as non-leaf nodes. Resource information is disseminated into each active router in ART, and nodes in ART work cooperatively to discover and schedule resources. Communications between domains are done via the root of ART. Second, the resource trade between consumers and providers is carried out by an improved barter marketing model. The proposed model reduces the complexity and cost of the trade, and provides desired soft quality of service as well. We illustrate that the proposed system exhibits good scalability, autonomy, flexibility, and fault-tolerance. 1 Introduction A grid [1] is a very large-scale network computing system that can scale up to Internet size environment, in which kinds of computing, storage and data resources, as well as scientific devices or implements, are distributed across multiple organizations and administrative domains. The computing resources can be supercomputers, SMPs, clusters, desktop PCs, or even mobile computing devices such as PDA. In the grid s architecture, grid resource management system (GRMS) is the central component, which is responsible for disseminating resources information across the grid, accepting requests for resources, discovering and scheduling the suitable resources that match the requests from the global grid resources, and executing the requests on scheduled resources [2]. As a grid is geographically distributed, heterogeneous and autonomous in nature, the design and implementation of RMS is challenging. In this paper, we focus on GRMS for computational grid. The scalability of GRMS determines the scalability of the grid, and the scalability of GRMS is determined by the resource organization structure, resource scheduling structure and economic model for resource trade. In this paper, we propose an active resource management system (AGRMS) supported by active networks for computa- * This research was supported by the National Natural Science Foundation of China under Grant No , the National High Technology Development 863 Program of China under Grant No.2001AA and the National Grand Fundamental Research 973 Program of China under Grant No.2002CB H. Jin, Y. Pan, N. Xiao, and J. Sun (Eds.): GCC 2004, LNCS 3251, pp , Springer-Verlag Berlin Heidelberg 2004

2 226 Xiaolin Chen et al. tional grid. AGRMS employs a two-level resource management architecture and an improved resource barter trade model, which demonstrate good scalability, autonomy, flexibility, extensibility and fault-tolerance, and provide soft quality of service. The remainder of this paper is organized as follows. In section 2 background and related works are reviewed. Section 3 presents the assumed application characteristics. The design of ARRMS is described in Section 4. Section 5 introduces the current status of the implementation. Conclusion and the future work are finally presented in Section 6. 2 Related Work Most GRMSs for computational grid, such as AppLeS [3], Condor [4], Globus [5], and Legion [6], lack an economic model for resource trade. Because the motivation for contributing resources to grid has been driven by public good, prize, fun or collaborative advantage, it makes these GRMSs not scale well. [7] presents a number of arguments for the need of market or economy-based mechanisms in GRMS. Economic Model for Resource trade can be classified into barter trade and trade with currency as mediate. The disadvantages of the former are that it is less flexible and it has limitation on single goods at a time, while the advantages of the latter is that it is flexible and it is suitable for multiple goods exchange, but its process is complicated and the cost of trade is too high, which are caused by the introduction of electronic cash, electronic bank or trade market due to pricing, accounting and payment mechanisms in the process of trade. Distributed computing systems that adopt the latter trade mode can t scale well, because there are some centralized electronic banks or trade markets existing to reduce the complication of trade. In this paper, we proposed an improved resource barter trade model for computational grid, which overcomes the disadvantages of traditional barter trade and reduces the complexity and cost of the trade. Active network [8, 9] is a novel approach to network architecture in which customized computation can be performed on the fly at routers as active packets pass through them. The user-defined computation carried by an active packet may extend and modify the network infrastructure. Active networks are composed of active nodes, which are either active routers or active terminal nodes (hosts). Active nodes provide execution environment (EE) for user-defined computation carried by the active packets. When an active packet that contains active code or references to code arrives at an active node, the code is extracted or retrieved and executed. The code can modify the content of the packet or the state of the active node, or transmit one or more packets. The active node also provides soft-state cache for the active code to store active data and other state information. 3 Basic Assumptions To focus our research on grid resource management, we made a simple assumption that applications running on AGRMS are long running SPMD (Single Program Multiple Data stream). In the SPMD model, each processor runs same program and operates on part of the overall data. There are many applications that are characteristic of SPMD model such as operation of two large-scale matrix, crypto attack by brute force, and so on. Furthermore we do not care for the issue of communication and

3 An Active Resource Management System for Computational Grid 227 collaboration between processes that run in parallel on multiple machines to implement the tasks of application. The assumption implies the following issues: Only computing resources are considered, so as to simplify resource description. We use a binary (C, M) (representing estimates of compute and memory) to describe resource provided by a machine at a certain moment. According to the computational capacity and time the application needs to run in a single machine, the user divides application data into chunks, and requests GRMS to schedule application code and data to multiple machines with resource chunk as a unit. A resource chunk is represented by a triple (Cr, Mr, Tr), which means it takes Tr scheduling intervals to finish the processing of the application data of the chunk with the computing speed of Cr and the memory of Mr. Because the type of resource needed is monotonous, only the CPU time and memory are used, it is suitable to adopt the barter trade model. For the convenience of calculating the amount of resource and guaranteeing the fairness of trade, we make a unified price standard, the price of 1MIPS per 1 scheduling interval is 1, and that of 1K memory per 1 scheduling interval is 1. 4 Active Grid Resource Management System 4.1 Architecture According to the behavior of autonomous system in the IP network, resources are divided into separate autonomous domains. In the domain, a hierarchical structure is constructed where the root node stands for the root of domain, the mediate node stands for sub-domain, and the leaf node stands for computing resources such as supercomputers, clusters, PCs, and etc. In the tree-structured domains, all the nodes are active routers (AR) except for the leaf nodes, root node can be the border router of IP autonomous system or not, and the connection between any two nodes can be either physical or logical. Thus, an active resource tree (ART) is built. Among domains, the root node is used to communicate with other domains, and circuits can exist in topology of domains. Resource information is disseminated to soft-state cache of each AR in ART, and the cooperation of nodes in ART realizes recourses discovery and schedule, thus building a hierarchical resource management structure within the domain. As the resource broker of each user in a domain, the root of ART (root node) trades on load and repayment with other domains. In the process of trading, the trade entity is domain but not the users of domain. Thus, a scalable two-level hierarchical architecture of resource management system is formed (see Fig.1). 4.2 Resource Dissemination In the domain, resource manager reports resource state within a week to resource active router (RAR) which neighbors to the resource by the way of resource state matrix RS[7][48]. Each element RS[i][j] of the matrix RS refers to the total number of available CPU and memory resource in a scheduling interval j in day i after today. A scheduling interval lasts for 30 minutes. In the last scheduling interval of each day, resource manager updates matrix RS which is stored in AR s soft-state cache. Each element of matrix RS is 8 bytes and the size of RS is less than 3k. Suppose that RAR

4 228 Xiaolin Chen et al. Fig. 1. The architecture of AGRMS. has n ports, and RS k stands for resource state matrix reported by port K. RAR computes the sub-domain s resource state matrix RS, n RS = ( K = RS k 1 ). (1) then reports the result to parent node. Similar to the RAR, each node reports to parent node about the sub-domain s RS except for root node. According to the number of child nodes and the soft-state cache s free space, AR can store each matrix RS k reported by each child node, or store its sub-domain s RS to save space. 4.3 Resource Trade Model Users are taken as trade entities of both sides in traditional barter trade model, which results in the following problems: In order to get the debt back, debtee may communicate with debtors that distributed on multiple machines of different domains. Demanding thousands of debtors to schedule the job at the same time is impossible, thus the deadline of completing the job cannot be guaranteed. Due to the great gap of the time to schedule the job of debtee among different debtors, the communication between processes running in parallel on multiple machines is complicated. This brings difficulty in designing grid application. In order to solve these problems, we use domain as trade entity to loan and repay resource among domains, and use user and domain as trade entities to loan and repay resource in the domain. Among different domains, root node is the resource agent of every user in the domain. It loans and repays resource from the outside for its clients, and maintains a debt database that records the transactions with other domains. When root node receives resource request from other domains, it will schedule the whole domain s idle resource to provide service; when root domain node receives repayment request from other domain, it will check the debt relation and schedule the whole domain s idle resources to repay. In the domain, each user maintains a balance which is initialized set to zero. When the user obtains resources from either inside or outside of the domain, it will subtract the price of resource from the balance, accordingly, when user provides resources for request which comes from either inside or outside of

5 An Active Resource Management System for Computational Grid 229 the domain, it will add the price of resource to the balance. For example, Alice and Mike are users of domain A. They borrow resources valued 100, 300 from domain B, then domain B schedules user John to provide resources valued 40, 100 to Alice and Mike respectively, and schedules user Martin to provide Alice and Mike resource valued 60, 200 respectively. Later, John from domain B requests resource valued 400 outside domain B, then domain B s root node asks domain A to pay back 400. Domain A schedules its user Alice, Mike and Merry to provide resource valued 100, 200, and 100 respectively to repay. If the balance of each above user is 0 at the beginning, when the trade ends, the balance of user Alice, Mike, Merry, John and Martin will be 0, -100, 100, -260 and 260 respectively. The process of trade is greatly simplified by making use of the methods aforementioned. Meanwhile the cost of communication and scheduling for realizing the trade is reduced, and a large amount of idle resource inside the domain can be scheduled by root node at a scheduling interval. This improves the system s efficiency and scalability; Moreover, it can provide soft quality of service. 4.4 Resource Discovery and Scheduling Resource Request The resource request sent from the user includes the following information: Rr (Cr, Rr, Tr): definition of a resource chunk. number: the number of resource chunks. balance: the current balance of requester. income: the average daily income that can be earned by requester within a week. deadline: the job requested should be completed before one scheduling internal ends. If there s no requirement for deadline, deadline is 0. bandwidth: minimum network bandwidth of resource provider. When receiving resource request, RAR forwards it to parent node. Nodes in ART forward the request to parent nodes until it reach to root node. As agent of user inside the domain, root node requests resource from inside and outside of the domain for its clients. If root node requests resource in the domain with a deadline, it computes the amount of resource each child node should sell according to resource state matrix RS k, and notifies each child node to sell a certain amount of resource between some scheduling interval and deadline. After the mediate node of ART receives notification of selling resource, the node does the same as root node does. When the user receives the above notification forwarded by RAR, it decides how much resource to be sold according to its own selling strategy. Meanwhile, it preserves resource and replies to RAR with the amount of resource to be sold and the updated resource state matrix RS k. Every user may independently designs its own selling strategy which defines how much resource to be sold according to such factors as risk coefficient, demand of network bandwidth and so on. Risk coefficient is defined as: Risk coefficient = (number * price( Rr ) balance) / income. (2) When receiving the responses for all the notifications, the mediate node of ART records the amount of resource to be sold by each child node in soft-state cache, and

6 230 Xiaolin Chen et al. reports to parent node about the total amount of resource to be sold and the updated resource state matrix of its child nodes. Finally, root node will know how much resource is actually sold to resource requester. If the requester is inside the domain, root node will inform the requester about the amount of resource to be sold. Otherwise, root node will calculate and record liquidated debt between the domain and the requester, and send the requester a notification which carries the information of the amount resource to be sold and the liquidated debt. After receiving the notification signed by resource provider, the requester signs an agreement about the liquidated debt and sends it to resource provider. If the resource requester send request without deadline, root node will put priority on scheduling users whose balance is negative to sell resource; if the total amount of resource provided by all such users cannot satisfy resource request, the remains should be processed according to the way of that of deadline, and the deadline is the last scheduling interval of that week. In order to implement schedule priority of debtors, each user reports to RAR about its balance. Each AR in ART maintains a debt bitmap in soft-state cache to indicate which child nodes have debtors Resource Repayment When receiving resource repayment notification from other domains, root node checks the validity of the notification base on the debt database. If the notification is not valid, root node demands the requester to send the last liquidated debt signed by root node to verify the sum root node owes. If the repayment requires a deadline, root node process in the way described in If there s no deadline, root node process in the similar way described in The difference is: in the latter case, the user can decide the amount of resource to be sold according to its own selling strategy after receiving selling resource notification; however, in the former case, the amount of resource to be sold is the maximum between the resource that is needed to repay off the debt and the resource that can be supplied within one week. When receiving resource request, root node first notify every debtors within the domain to repay to avoid that some debtors can t earn money for a long time and can t ask for loans Resource Scheduling After receiving response for resource request from root node, resource requester will send active codes and data which realize computing task to root node. Root node will distribute the active codes and data to child nodes according to the amount of resource preserved by child nodes, and actions of the rest mediate node in ART is the same as root node s. After receiving active code and application data, resource provider will executes active code on assigned scheduling intervals Resource Authorization In the domain, any sub-domain can authorize a user of another sub-domain to consume its resource free or on discount. And it is the same case between domains. If some domain or sub-domain authorizes a user of another sub-domains to consume its resource free or on discount, its administrator will send authorization statements to active router which charges that domain. The active router will store these statements in soft-state cache.

7 An Active Resource Management System for Computational Grid Fault-Tolerance When root node fails, a new root node will be selected from its child nodes. After logically connecting all child node of the old root, the new root node asks them to report resource information matrix RS, debtor bitmap and so on. It also requests resource provider to report records of trade with other domains. After receiving the response, the new root node computes out debt relation between its domain and other domain. Thus, the state of the new root is the same as that of the old one before it fails. When the mediate node fails, process in the similar way as the previous one does. In addition, root node can store redundant resource proportion in soft-state cache. When some resource provider or mediate node fail, the nodes which detect these failures will report to root node about the amount of resource planned to be provided by these failed nodes. Root node allocates corresponding resource from redundant resource. 5 Implementation Status We are currently implementing AGRMS on the ANTS [10] toolkit. ANTS is a javabased active network execution environment developed by MIT. It has three components: Capsule, Active Node and Code distribution System. A capsule encapsulates the application data and user customized program. When the capsules pass through active nodes that can be routers, switches or end system, user-defined codes carried in the capsule are executed and routed to destination. Code distribution System provides on-demand code distribution using mobile code. Most components of AGRMS are implemented by ten types of capsule. Some components have been finished or nearing completion. Other parts remain being developed. 6 Conclusions and Future Work In this paper, we propose an active grid resource management system (AGRMS) supported by active networks for computational grid. AGRMS employs a two-level resource management architecture and an improved resource barter trade model, which demonstrate good scalability, autonomy, flexibility, extensibility and faulttolerance, and provide soft quality of service. To simplify the model and concentrate on resource management strategy, we have not discussed the communication between processes distributed on multiple machines within a grid in this paper. As a supplement, we are doing some research work from the following two aspects. First, we are designing a layered process communication service based on MPI (Message Passing Interface) to accommodate layered scheduling. Second, network bandwidth will be further considered in the resource model, and AGRMS will guarantee soft qualify of service also on bandwidth. References 1. I. Foster and C. Kesselman (editors), The Grid: Blueprint for a New Computing Infrastructure, Morgan Kaufmann Publishers, USA, Klaus Krauter, et al, A Taxonomy and Survey of Grid Resource Management Systems, International Journal of Software, Practice and Experience, Volume 32, Issue 2, Pages: , Wiley Press, USA, 2002.

8 232 Xiaolin Chen et al. 3. F. Berman and R. Wolski, The AppLeS Project: A Status Report, Proceedings of the Eight NEC Research Symposium, Germany, May M. Litzkow, et al, Condor - A Hunter of Idle Workstations, Proceedings of the 8th International Conference of Distributed Computing Systems, June I. Foster and C. Kesselman., Globus: A Metacomputing Infrastructure Toolkit, International Journal of Supercomputer Applications, 11(2): , S. Chapin, et al, The Legion Resource Management System, Proceedings of the 5th Workshop on Job Scheduling Strategies for Parallel Processing, April R. Buyya, et al D. Abramson, J. Giddy, An Economy Driven Resource Management Architecture for Global Computational Power Grids, The 2000 International Conference on Parallel and Distributed Processing Techniques and Applications (PDPTA 2000), Las Vegas, USA, June 26-29, D.L.Tenenhouse and D.Wetherall, Towards an Active Network Architecture, Proc. Multimedia Comp and Networking 96, MMCN' 96, San Jose, CA, Jan D.L.Tenenouse et al, A Survey of Active Network Research, IEEE Communications Manazine, 1997, 35(1): D.Wetherall et al, ANTS: A toolkit for building and dynamically depoloying network protocols, Proc. IEEE OpenArch' 98[C],San Francisco, CA, Apr,1998.

A Resource Discovery Algorithm in Mobile Grid Computing Based on IP-Paging Scheme

A Resource Discovery Algorithm in Mobile Grid Computing Based on IP-Paging Scheme A Resource Discovery Algorithm in Mobile Grid Computing Based on IP-Paging Scheme Yue Zhang 1 and Yunxia Pei 2 1 Department of Math and Computer Science Center of Network, Henan Police College, Zhengzhou,

More information

A Resource Discovery Algorithm in Mobile Grid Computing based on IP-paging Scheme

A Resource Discovery Algorithm in Mobile Grid Computing based on IP-paging Scheme A Resource Discovery Algorithm in Mobile Grid Computing based on IP-paging Scheme Yue Zhang, Yunxia Pei To cite this version: Yue Zhang, Yunxia Pei. A Resource Discovery Algorithm in Mobile Grid Computing

More information

Grid Computing Systems: A Survey and Taxonomy

Grid Computing Systems: A Survey and Taxonomy Grid Computing Systems: A Survey and Taxonomy Material for this lecture from: A Survey and Taxonomy of Resource Management Systems for Grid Computing Systems, K. Krauter, R. Buyya, M. Maheswaran, CS Technical

More information

GRB. Grid-JQA : Grid Java based Quality of service management by Active database. L. Mohammad Khanli M. Analoui. Abstract.

GRB. Grid-JQA : Grid Java based Quality of service management by Active database. L. Mohammad Khanli M. Analoui. Abstract. Grid-JQA : Grid Java based Quality of service management by Active database L. Mohammad Khanli M. Analoui Ph.D. student C.E. Dept. IUST Tehran, Iran Khanli@iust.ac.ir Assistant professor C.E. Dept. IUST

More information

THE VEGA PERSONAL GRID: A LIGHTWEIGHT GRID ARCHITECTURE

THE VEGA PERSONAL GRID: A LIGHTWEIGHT GRID ARCHITECTURE THE VEGA PERSONAL GRID: A LIGHTWEIGHT GRID ARCHITECTURE Wei Li, Zhiwei Xu, Bingchen Li, Yili Gong Institute of Computing Technology of Chinese Academy of Sciences Beijing China, 100080 {zxu, liwei, libingchen,

More information

Grid Computing. Grid Computing 2

Grid Computing. Grid Computing 2 Grid Computing Mahesh Joshi joshi031@d.umn.edu Presentation for Graduate Course in Advanced Computer Architecture 28 th April 2005 Objective Overview of the concept and related aspects Some practical implications

More information

MSF: A Workflow Service Infrastructure for Computational Grid Environments

MSF: A Workflow Service Infrastructure for Computational Grid Environments MSF: A Workflow Service Infrastructure for Computational Grid Environments Seogchan Hwang 1 and Jaeyoung Choi 2 1 Supercomputing Center, Korea Institute of Science and Technology Information, 52 Eoeun-dong,

More information

A Finite State Mobile Agent Computation Model

A Finite State Mobile Agent Computation Model A Finite State Mobile Agent Computation Model Yong Liu, Congfu Xu, Zhaohui Wu, Weidong Chen, and Yunhe Pan College of Computer Science, Zhejiang University Hangzhou 310027, PR China Abstract In this paper,

More information

Simulating a Finite State Mobile Agent System

Simulating a Finite State Mobile Agent System Simulating a Finite State Mobile Agent System Liu Yong, Xu Congfu, Chen Yanyu, and Pan Yunhe College of Computer Science, Zhejiang University, Hangzhou 310027, P.R. China Abstract. This paper analyzes

More information

Scheduling Large Parametric Modelling Experiments on a Distributed Meta-computer

Scheduling Large Parametric Modelling Experiments on a Distributed Meta-computer Scheduling Large Parametric Modelling Experiments on a Distributed Meta-computer David Abramson and Jon Giddy Department of Digital Systems, CRC for Distributed Systems Technology Monash University, Gehrmann

More information

A Chord-Based Novel Mobile Peer-to-Peer File Sharing Protocol

A Chord-Based Novel Mobile Peer-to-Peer File Sharing Protocol A Chord-Based Novel Mobile Peer-to-Peer File Sharing Protocol Min Li 1, Enhong Chen 1, and Phillip C-y Sheu 2 1 Department of Computer Science and Technology, University of Science and Technology of China,

More information

CS Advanced Topics in Database Management Systems. Report: P2P vs Grid Computing. By Ooi Hong Sain HT029804y

CS Advanced Topics in Database Management Systems. Report: P2P vs Grid Computing. By Ooi Hong Sain HT029804y CS 6203 Advanced Topics in Database Management Systems Report: P2P vs Grid Computing By Ooi Hong Sain HT029804y Acknowledgement All figures are taken directly from the original works [buyya2000, buyya2002].

More information

A Data-Aware Resource Broker for Data Grids

A Data-Aware Resource Broker for Data Grids A Data-Aware Resource Broker for Data Grids Huy Le, Paul Coddington, and Andrew L. Wendelborn School of Computer Science, University of Adelaide Adelaide, SA 5005, Australia {paulc,andrew}@cs.adelaide.edu.au

More information

A Time-To-Live Based Reservation Algorithm on Fully Decentralized Resource Discovery in Grid Computing

A Time-To-Live Based Reservation Algorithm on Fully Decentralized Resource Discovery in Grid Computing A Time-To-Live Based Reservation Algorithm on Fully Decentralized Resource Discovery in Grid Computing Sanya Tangpongprasit, Takahiro Katagiri, Hiroki Honda, Toshitsugu Yuba Graduate School of Information

More information

A Grid-Enabled Component Container for CORBA Lightweight Components

A Grid-Enabled Component Container for CORBA Lightweight Components A Grid-Enabled Component Container for CORBA Lightweight Components Diego Sevilla 1, José M. García 1, Antonio F. Gómez 2 1 Department of Computer Engineering 2 Department of Information and Communications

More information

A Heuristic Based Load Balancing Algorithm

A Heuristic Based Load Balancing Algorithm International Journal of Computational Engineering & Management, Vol. 15 Issue 6, November 2012 www..org 56 A Heuristic Based Load Balancing Algorithm 1 Harish Rohil, 2 Sanjna Kalyan 1,2 Department of

More information

An Engineering Computation Oriented Visual Grid Framework

An Engineering Computation Oriented Visual Grid Framework An Engineering Computation Oriented Visual Grid Framework Guiyi Wei 1,2,3, Yao Zheng 1,2, Jifa Zhang 1,2, and Guanghua Song 1,2 1 College of Computer Science, Zhejiang University, Hangzhou, 310027, P.

More information

Grid Resources Search Engine based on Ontology

Grid Resources Search Engine based on Ontology based on Ontology 12 E-mail: emiao_beyond@163.com Yang Li 3 E-mail: miipl606@163.com Weiguang Xu E-mail: miipl606@163.com Jiabao Wang E-mail: miipl606@163.com Lei Song E-mail: songlei@nudt.edu.cn Jiang

More information

A Mobile Agent-based Model for Service Management in Virtual Active Networks

A Mobile Agent-based Model for Service Management in Virtual Active Networks A Mobile Agent-based Model for Service Management in Virtual Active Networks Fábio Luciano Verdi and Edmundo R. M. Madeira Institute of Computing, University of Campinas (UNICAMP), Campinas-SP, Brazil

More information

Two-Level Dynamic Load Balancing Algorithm Using Load Thresholds and Pairwise Immigration

Two-Level Dynamic Load Balancing Algorithm Using Load Thresholds and Pairwise Immigration Two-Level Dynamic Load Balancing Algorithm Using Load Thresholds and Pairwise Immigration Hojiev Sardor Qurbonboyevich Department of IT Convergence Engineering Kumoh National Institute of Technology, Daehak-ro

More information

Resource CoAllocation for Scheduling Tasks with Dependencies, in Grid

Resource CoAllocation for Scheduling Tasks with Dependencies, in Grid Resource CoAllocation for Scheduling Tasks with Dependencies, in Grid Diana Moise 1,2, Izabela Moise 1,2, Florin Pop 1, Valentin Cristea 1 1 University Politehnica of Bucharest, Romania 2 INRIA/IRISA,

More information

Application of Redundant Backup Technology in Network Security

Application of Redundant Backup Technology in Network Security 2018 2nd International Conference on Systems, Computing, and Applications (SYSTCA 2018) Application of Redundant Backup Technology in Network Security Shuwen Deng1, Siping Hu*, 1, Dianhua Wang1, Limin

More information

Visual Modeler for Grid Modeling and Simulation (GridSim) Toolkit

Visual Modeler for Grid Modeling and Simulation (GridSim) Toolkit Visual Modeler for Grid Modeling and Simulation (GridSim) Toolkit Anthony Sulistio, Chee Shin Yeo, and Rajkumar Buyya Grid Computing and Distributed Systems (GRIDS) Laboratory, Department of Computer Science

More information

Assignment 5. Georgia Koloniari

Assignment 5. Georgia Koloniari Assignment 5 Georgia Koloniari 2. "Peer-to-Peer Computing" 1. What is the definition of a p2p system given by the authors in sec 1? Compare it with at least one of the definitions surveyed in the last

More information

Proffering a new Method for Grid Computing Resource Discovery based on Economic Criteria using Ant Colony Algorithm

Proffering a new Method for Grid Computing Resource Discovery based on Economic Criteria using Ant Colony Algorithm Volume 3 No., September 2 Proffering a new Method for Grid Computing Resource Discovery based on Economic Criteria using Ant Colony Algorithm Ali Sarhadi Department of Computer Engineering, Toyserkan Branch,

More information

Exploring the Catallactic Coordination Approach for Peer-to-Peer systems *

Exploring the Catallactic Coordination Approach for Peer-to-Peer systems * Exploring the Catallactic Coordination Approach for Peer-to-Peer systems * Oscar Ardaiz 1, Pau Artigas 1, Torsten Eymann 2, Felix Freitag 1, Roc Messeguer 1, Leandro Navarro 1, Michael Reinicke 2 1 Computer

More information

Scalable Hybrid Search on Distributed Databases

Scalable Hybrid Search on Distributed Databases Scalable Hybrid Search on Distributed Databases Jungkee Kim 1,2 and Geoffrey Fox 2 1 Department of Computer Science, Florida State University, Tallahassee FL 32306, U.S.A., jungkkim@cs.fsu.edu, 2 Community

More information

Framework for Preventing Deadlock : A Resource Co-allocation Issue in Grid Environment

Framework for Preventing Deadlock : A Resource Co-allocation Issue in Grid Environment Framework for Preventing Deadlock : A Resource Co-allocation Issue in Grid Environment Dr. Deepti Malhotra Department of Computer Science and Information Technology Central University of Jammu, Jammu,

More information

Introduction to GT3. Introduction to GT3. What is a Grid? A Story of Evolution. The Globus Project

Introduction to GT3. Introduction to GT3. What is a Grid? A Story of Evolution. The Globus Project Introduction to GT3 The Globus Project Argonne National Laboratory USC Information Sciences Institute Copyright (C) 2003 University of Chicago and The University of Southern California. All Rights Reserved.

More information

Outline. Definition of a Distributed System Goals of a Distributed System Types of Distributed Systems

Outline. Definition of a Distributed System Goals of a Distributed System Types of Distributed Systems Distributed Systems Outline Definition of a Distributed System Goals of a Distributed System Types of Distributed Systems What Is A Distributed System? A collection of independent computers that appears

More information

A Resource Look up Strategy for Distributed Computing

A Resource Look up Strategy for Distributed Computing A Resource Look up Strategy for Distributed Computing F. AGOSTARO, A. GENCO, S. SORCE DINFO - Dipartimento di Ingegneria Informatica Università degli Studi di Palermo Viale delle Scienze, edificio 6 90128

More information

MANAGEMENT AND PLACEMENT OF REPLICAS IN A HIERARCHICAL DATA GRID

MANAGEMENT AND PLACEMENT OF REPLICAS IN A HIERARCHICAL DATA GRID MANAGEMENT AND PLACEMENT OF REPLICAS IN A HIERARCHICAL DATA GRID Ghalem Belalem 1 and Bakhta Meroufel 2 1 Department of Computer Science, Faculty of Sciences, University of Oran (Es Senia), Algeria ghalem1dz@gmail.com

More information

CloudSim: A Toolkit for Modeling and Simulation of Cloud Computing Environments

CloudSim: A Toolkit for Modeling and Simulation of Cloud Computing Environments CloudSim: A Toolkit for Modeling and Simulation of Cloud Computing Environments Presented by: Dr. Faramarz Safi Islamic Azad University, Najafabad Branch, Esfahan, Iran. and with special thanks to Mrs.

More information

Evaluating Algorithms for Shared File Pointer Operations in MPI I/O

Evaluating Algorithms for Shared File Pointer Operations in MPI I/O Evaluating Algorithms for Shared File Pointer Operations in MPI I/O Ketan Kulkarni and Edgar Gabriel Parallel Software Technologies Laboratory, Department of Computer Science, University of Houston {knkulkarni,gabriel}@cs.uh.edu

More information

Functional Requirements for Grid Oriented Optical Networks

Functional Requirements for Grid Oriented Optical Networks Functional Requirements for Grid Oriented Optical s Luca Valcarenghi Internal Workshop 4 on Photonic s and Technologies Scuola Superiore Sant Anna Pisa June 3-4, 2003 1 Motivations Grid networking connection

More information

Design of Distributed Data Mining Applications on the KNOWLEDGE GRID

Design of Distributed Data Mining Applications on the KNOWLEDGE GRID Design of Distributed Data Mining Applications on the KNOWLEDGE GRID Mario Cannataro ICAR-CNR cannataro@acm.org Domenico Talia DEIS University of Calabria talia@deis.unical.it Paolo Trunfio DEIS University

More information

Text mining on a grid environment

Text mining on a grid environment Data Mining X 13 Text mining on a grid environment V. G. Roncero, M. C. A. Costa & N. F. F. Ebecken COPPE/Federal University of Rio de Janeiro, Brazil Abstract The enormous amount of information stored

More information

what do we mean by event processing now, a checklist of capabilities in current event processing tools and applications,

what do we mean by event processing now, a checklist of capabilities in current event processing tools and applications, A View of the Current State of Event Processing what do we mean by event processing now, complex event processing, a checklist of capabilities in current event processing tools and applications, next steps

More information

A Decoupled Scheduling Approach for the GrADS Program Development Environment. DCSL Ahmed Amin

A Decoupled Scheduling Approach for the GrADS Program Development Environment. DCSL Ahmed Amin A Decoupled Scheduling Approach for the GrADS Program Development Environment DCSL Ahmed Amin Outline Introduction Related Work Scheduling Architecture Scheduling Algorithm Testbench Results Conclusions

More information

Client-server Basics. 1. Motivation

Client-server Basics. 1. Motivation Page 1 of 8 Client-server Basics Oxford Brookes University 2004 Contents 1. Motivation 2. Basic Model of the Web 3. Extensions - Caching and Proxies 4. High Performance Web Servers 5. Web Services Appendices

More information

The power quality intelligent monitoring system based on cloud computing Jie Bai 1a, Changpo Song 2b

The power quality intelligent monitoring system based on cloud computing Jie Bai 1a, Changpo Song 2b International Conference on Intelligent Systems Research and Mechatronics Engineering (ISRME 2015) The power quality intelligent monitoring system based on cloud computing Jie Bai 1a, Changpo Song 2b State

More information

Clustering-Based Distributed Precomputation for Quality-of-Service Routing*

Clustering-Based Distributed Precomputation for Quality-of-Service Routing* Clustering-Based Distributed Precomputation for Quality-of-Service Routing* Yong Cui and Jianping Wu Department of Computer Science, Tsinghua University, Beijing, P.R.China, 100084 cy@csnet1.cs.tsinghua.edu.cn,

More information

Introduction to Grid Computing

Introduction to Grid Computing Milestone 2 Include the names of the papers You only have a page be selective about what you include Be specific; summarize the authors contributions, not just what the paper is about. You might be able

More information

UNICORE Globus: Interoperability of Grid Infrastructures

UNICORE Globus: Interoperability of Grid Infrastructures UNICORE : Interoperability of Grid Infrastructures Michael Rambadt Philipp Wieder Central Institute for Applied Mathematics (ZAM) Research Centre Juelich D 52425 Juelich, Germany Phone: +49 2461 612057

More information

HETEROGENEOUS COMPUTING

HETEROGENEOUS COMPUTING HETEROGENEOUS COMPUTING Shoukat Ali, Tracy D. Braun, Howard Jay Siegel, and Anthony A. Maciejewski School of Electrical and Computer Engineering, Purdue University Heterogeneous computing is a set of techniques

More information

Online Optimization of VM Deployment in IaaS Cloud

Online Optimization of VM Deployment in IaaS Cloud Online Optimization of VM Deployment in IaaS Cloud Pei Fan, Zhenbang Chen, Ji Wang School of Computer Science National University of Defense Technology Changsha, 4173, P.R.China {peifan,zbchen}@nudt.edu.cn,

More information

A Compact Computing Environment For A Windows PC Cluster Towards Seamless Molecular Dynamics Simulations

A Compact Computing Environment For A Windows PC Cluster Towards Seamless Molecular Dynamics Simulations A Compact Computing Environment For A Windows PC Cluster Towards Seamless Molecular Dynamics Simulations Yuichi Tsujita Abstract A Windows PC cluster is focused for its high availabilities and fruitful

More information

Resource Load Balancing Based on Multi-agent in ServiceBSP Model*

Resource Load Balancing Based on Multi-agent in ServiceBSP Model* Resource Load Balancing Based on Multi-agent in ServiceBSP Model* Yan Jiang 1, Weiqin Tong 1, and Wentao Zhao 2 1 School of Computer Engineering and Science, Shanghai University 2 Image Processing and

More information

GRID COMPUTING BASED MODEL FOR REMOTE MONITORING OF ENERGY FLOW AND PREDICTION OF HT LINE LOSS IN POWER DISTRIBUTION SYSTEM

GRID COMPUTING BASED MODEL FOR REMOTE MONITORING OF ENERGY FLOW AND PREDICTION OF HT LINE LOSS IN POWER DISTRIBUTION SYSTEM GRID COMPUTING BASED MODEL FOR REMOTE MONITORING OF ENERGY FLOW AND PREDICTION OF HT LINE LOSS IN POWER DISTRIBUTION SYSTEM 1 C.Senthamarai, 2 A.Krishnan 1 Assistant Professor., Department of MCA, K.S.Rangasamy

More information

Global IP Network System Large-Scale, Guaranteed, Carrier-Grade

Global IP Network System Large-Scale, Guaranteed, Carrier-Grade Global Network System Large-Scale, Guaranteed, Carrier-Grade 192 Global Network System Large-Scale, Guaranteed, Carrier-Grade Takanori Miyamoto Shiro Tanabe Osamu Takada Shinobu Gohara OVERVIEW: traffic

More information

System models for distributed systems

System models for distributed systems System models for distributed systems INF5040/9040 autumn 2010 lecturer: Frank Eliassen INF5040 H2010, Frank Eliassen 1 System models Purpose illustrate/describe common properties and design choices for

More information

Load Balancing Algorithm over a Distributed Cloud Network

Load Balancing Algorithm over a Distributed Cloud Network Load Balancing Algorithm over a Distributed Cloud Network Priyank Singhal Student, Computer Department Sumiran Shah Student, Computer Department Pranit Kalantri Student, Electronics Department Abstract

More information

Computational Mini-Grid Research at Clemson University

Computational Mini-Grid Research at Clemson University Computational Mini-Grid Research at Clemson University Parallel Architecture Research Lab November 19, 2002 Project Description The concept of grid computing is becoming a more and more important one in

More information

QoS Guided Min-Mean Task Scheduling Algorithm for Scheduling Dr.G.K.Kamalam

QoS Guided Min-Mean Task Scheduling Algorithm for Scheduling Dr.G.K.Kamalam International Journal of Computer Communication and Information System(IJJCCIS) Vol 7. No.1 215 Pp. 1-7 gopalax Journals, Singapore available at : www.ijcns.com ISSN: 976 1349 ---------------------------------------------------------------------------------------------------------------------

More information

Job Re-Packing for Enhancing the Performance of Gang Scheduling

Job Re-Packing for Enhancing the Performance of Gang Scheduling Job Re-Packing for Enhancing the Performance of Gang Scheduling B. B. Zhou 1, R. P. Brent 2, C. W. Johnson 3, and D. Walsh 3 1 Computer Sciences Laboratory, Australian National University, Canberra, ACT

More information

Hungarian Supercomputing Grid 1

Hungarian Supercomputing Grid 1 Hungarian Supercomputing Grid 1 Péter Kacsuk MTA SZTAKI Victor Hugo u. 18-22, Budapest, HUNGARY www.lpds.sztaki.hu E-mail: kacsuk@sztaki.hu Abstract. The main objective of the paper is to describe the

More information

An Architecture For Computational Grids Based On Proxy Servers

An Architecture For Computational Grids Based On Proxy Servers An Architecture For Computational Grids Based On Proxy Servers P. V. C. Costa, S. D. Zorzo, H. C. Guardia {paulocosta,zorzo,helio}@dc.ufscar.br UFSCar Federal University of São Carlos, Brazil Abstract

More information

Performance Analysis of Applying Replica Selection Technology for Data Grid Environments*

Performance Analysis of Applying Replica Selection Technology for Data Grid Environments* Performance Analysis of Applying Replica Selection Technology for Data Grid Environments* Chao-Tung Yang 1,, Chun-Hsiang Chen 1, Kuan-Ching Li 2, and Ching-Hsien Hsu 3 1 High-Performance Computing Laboratory,

More information

The ESB dynamic routing strategy in the low bandwidth network environment

The ESB dynamic routing strategy in the low bandwidth network environment Journal of Network Computing and Applications (2016) 1: 26-32 Clausius Scientific Press, Canada The ESB dynamic routing strategy in the low bandwidth network environment Wei Huang1,a, Kangyi Luo1, Baocheng

More information

6LPXODWLRQÃRIÃWKHÃ&RPPXQLFDWLRQÃ7LPHÃIRUÃDÃ6SDFH7LPH $GDSWLYHÃ3URFHVVLQJÃ$OJRULWKPÃRQÃDÃ3DUDOOHOÃ(PEHGGHG 6\VWHP

6LPXODWLRQÃRIÃWKHÃ&RPPXQLFDWLRQÃ7LPHÃIRUÃDÃ6SDFH7LPH $GDSWLYHÃ3URFHVVLQJÃ$OJRULWKPÃRQÃDÃ3DUDOOHOÃ(PEHGGHG 6\VWHP LPXODWLRQÃRIÃWKHÃ&RPPXQLFDWLRQÃLPHÃIRUÃDÃSDFHLPH $GDSWLYHÃURFHVVLQJÃ$OJRULWKPÃRQÃDÃDUDOOHOÃ(PEHGGHG \VWHP Jack M. West and John K. Antonio Department of Computer Science, P.O. Box, Texas Tech University,

More information

WARMStones: Benchmarking. Wide-Area Resource Management Schedulers

WARMStones: Benchmarking. Wide-Area Resource Management Schedulers WARMStones: Benchmarking Wide-Area Resource Management Schedulers Steve J. Chapin University of Virginia DRAFT White Paper Abstract Researchers have proposed hundreds of algorithms for solutions to the

More information

Distributed Scheduling for the Sombrero Single Address Space Distributed Operating System

Distributed Scheduling for the Sombrero Single Address Space Distributed Operating System Distributed Scheduling for the Sombrero Single Address Space Distributed Operating System Donald S. Miller Department of Computer Science and Engineering Arizona State University Tempe, AZ, USA Alan C.

More information

A Fast Handover Protocol for Mobile IPv6 Using Mobility Prediction Mechanism

A Fast Handover Protocol for Mobile IPv6 Using Mobility Prediction Mechanism A Fast Handover Protocol for Mobile IPv6 Using Mobility Prediction Mechanism Dae Sun Kim 1 and Choong Seon Hong 2 1 School of Electronics and Information, Kyung Hee Univerity 1 Seocheon, Giheung, Yongin,

More information

Top-down definition of Network Centric Operating System features

Top-down definition of Network Centric Operating System features Position paper submitted to the Workshop on Network Centric Operating Systems Bruxelles 16-17 march 2005 Top-down definition of Network Centric Operating System features Thesis Marco Danelutto Dept. Computer

More information

SDS: A Scalable Data Services System in Data Grid

SDS: A Scalable Data Services System in Data Grid SDS: A Scalable Data s System in Data Grid Xiaoning Peng School of Information Science & Engineering, Central South University Changsha 410083, China Department of Computer Science and Technology, Huaihua

More information

Resolving Load Balancing Issue of Grid Computing through Dynamic Approach

Resolving Load Balancing Issue of Grid Computing through Dynamic Approach Resolving Load Balancing Issue of Grid Computing through Dynamic Er. Roma Soni M-Tech Student Dr. Kamal Sharma Prof. & Director of E.C.E. Deptt. EMGOI, Badhauli. Er. Sharad Chauhan Asst. Prof. in C.S.E.

More information

visperf: Monitoring Tool for Grid Computing

visperf: Monitoring Tool for Grid Computing visperf: Monitoring Tool for Grid Computing DongWoo Lee 1, Jack J. Dongarra 2, and R.S. Ramakrishna 1 1 Department of Information and Communication Kwangju Institute of Science and Technology, Republic

More information

A Distributed Media Service System Based on Globus Data-Management Technologies1

A Distributed Media Service System Based on Globus Data-Management Technologies1 A Distributed Media Service System Based on Globus Data-Management Technologies1 Xiang Yu, Shoubao Yang, and Yu Hong Dept. of Computer Science, University of Science and Technology of China, Hefei 230026,

More information

System Models. 2.1 Introduction 2.2 Architectural Models 2.3 Fundamental Models. Nicola Dragoni Embedded Systems Engineering DTU Informatics

System Models. 2.1 Introduction 2.2 Architectural Models 2.3 Fundamental Models. Nicola Dragoni Embedded Systems Engineering DTU Informatics System Models Nicola Dragoni Embedded Systems Engineering DTU Informatics 2.1 Introduction 2.2 Architectural Models 2.3 Fundamental Models Architectural vs Fundamental Models Systems that are intended

More information

A Comparative Study on Exact Triangle Counting Algorithms on the GPU

A Comparative Study on Exact Triangle Counting Algorithms on the GPU A Comparative Study on Exact Triangle Counting Algorithms on the GPU Leyuan Wang, Yangzihao Wang, Carl Yang, John D. Owens University of California, Davis, CA, USA 31 st May 2016 L. Wang, Y. Wang, C. Yang,

More information

The main purpose of load balancing is to

The main purpose of load balancing is to INTERNATIONAL JOURNAL OF NETWORK MANAGEMENT Int. J. Network Mgmt 2005; 15: 311 319 Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/nem.567 IP layer load balance using

More information

Grids of Agents for Computer and Telecommunication Network Management

Grids of Agents for Computer and Telecommunication Network Management Grids of Agents for Computer and Telecommunication Network Marcos Dias de Assunção, Carlos Becker Westphall Network and Laboratory Federal University of Santa Catarina Florianópolis, SC, 88049-970, PO

More information

Chapter 4:- Introduction to Grid and its Evolution. Prepared By:- NITIN PANDYA Assistant Professor SVBIT.

Chapter 4:- Introduction to Grid and its Evolution. Prepared By:- NITIN PANDYA Assistant Professor SVBIT. Chapter 4:- Introduction to Grid and its Evolution Prepared By:- Assistant Professor SVBIT. Overview Background: What is the Grid? Related technologies Grid applications Communities Grid Tools Case Studies

More information

An Experience in Accessing Grid Computing from Mobile Device with GridLab Mobile Services

An Experience in Accessing Grid Computing from Mobile Device with GridLab Mobile Services An Experience in Accessing Grid Computing from Mobile Device with GridLab Mobile Services Riri Fitri Sari, Rene Paulus Department of Electrical Engineering, Faculty of Engineering University of Indonesia

More information

Hierarchical Disk Driven Scheduling for Mixed Workloads

Hierarchical Disk Driven Scheduling for Mixed Workloads Hierarchical Disk Driven Scheduling for Mixed Workloads Gwendolyn Einfeld gwen@soe.ucsc.edu Computer Science Department University of California, Santa Cruz Santa Cruz, CA 9564 Abstract Disk schedulers

More information

Design of Labour Agency Platform Based on Agent Technology of JADE *

Design of Labour Agency Platform Based on Agent Technology of JADE * Design of Labour Agency Platform Based on Agent Technology of JADE * Xiaobin Qiu **, Nan Zhou, and Xin Wang Network Center, China Agriculture University, Beijing 100083, P.R. China qxb@cau.edu.cn Abstract.

More information

InteGrade: a Tool for Executing Parallel Applications on a Grid for Opportunistic Computing

InteGrade: a Tool for Executing Parallel Applications on a Grid for Opportunistic Computing InteGrade: a Tool for Executing Parallel Applications on a Grid for Opportunistic Computing Jose de R. B. Pinheiro Junior, Raphael Y. de Camargo, Andrei Goldchleger, Fabio Kon 1 Department of Computer

More information

Crises Management in Multiagent Workflow Systems

Crises Management in Multiagent Workflow Systems Crises Management in Multiagent Workflow Systems Małgorzata Żabińska Department of Computer Science, AGH University of Science and Technology, al. Mickiewicza 30, 30-059 Kraków, Poland zabinska@agh.edu.pl

More information

Tree-Based Minimization of TCAM Entries for Packet Classification

Tree-Based Minimization of TCAM Entries for Packet Classification Tree-Based Minimization of TCAM Entries for Packet Classification YanSunandMinSikKim School of Electrical Engineering and Computer Science Washington State University Pullman, Washington 99164-2752, U.S.A.

More information

A flexible router platform for next generation network services

A flexible router platform for next generation network services A flexible router platform for next generation network services Lukas Ruf, Arno Wagner, Karoly Farkas, Bernhard Plattner Computer Engineering and Networks Laboratory (TIK) Swiss Federal Institute of Technology

More information

A Scalable Location Aware Service Platform for Mobile Applications Based on Java RMI

A Scalable Location Aware Service Platform for Mobile Applications Based on Java RMI A Scalable Location Aware Service Platform for Mobile Applications Based on Java RMI Olaf Droegehorn, Kirti Singh-Kurbel, Markus Franz, Roland Sorge, Rita Winkler, and Klaus David IHP, Im Technologiepark

More information

Performance of DB2 Enterprise-Extended Edition on NT with Virtual Interface Architecture

Performance of DB2 Enterprise-Extended Edition on NT with Virtual Interface Architecture Performance of DB2 Enterprise-Extended Edition on NT with Virtual Interface Architecture Sivakumar Harinath 1, Robert L. Grossman 1, K. Bernhard Schiefer 2, Xun Xue 2, and Sadique Syed 2 1 Laboratory of

More information

International Journal of Scientific & Engineering Research Volume 8, Issue 5, May ISSN

International Journal of Scientific & Engineering Research Volume 8, Issue 5, May ISSN International Journal of Scientific & Engineering Research Volume 8, Issue 5, May-2017 106 Self-organizing behavior of Wireless Ad Hoc Networks T. Raghu Trivedi, S. Giri Nath Abstract Self-organization

More information

A Comparison of Conventional Distributed Computing Environments and Computational Grids

A Comparison of Conventional Distributed Computing Environments and Computational Grids A Comparison of Conventional Distributed Computing Environments and Computational Grids Zsolt Németh 1, Vaidy Sunderam 2 1 MTA SZTAKI, Computer and Automation Research Institute, Hungarian Academy of Sciences,

More information

Simulation of a cost model response requests for replication in data grid environment

Simulation of a cost model response requests for replication in data grid environment Simulation of a cost model response requests for replication in data grid environment Benatiallah ali, Kaddi mohammed, Benatiallah djelloul, Harrouz abdelkader Laboratoire LEESI, faculté des science et

More information

Interoperable and Transparent Dynamic Deployment of Web Services for Service Oriented Grids

Interoperable and Transparent Dynamic Deployment of Web Services for Service Oriented Grids Interoperable and Transparent Dynamic Deployment of Web s for Oriented Grids Michael Messig and Andrzej Goscinski School of Engineering and Information Technology Deakin University Pigdons Road, Geelong

More information

NIRA: A New Internet Routing Architecture

NIRA: A New Internet Routing Architecture NIRA: A New Internet Routing Architecture Xiaowei Yang MIT Computer Science and Artificial Intelligence Laboratory yxw@lcs.mit.edu 1 Why a New Internet Routing Architecture? Users have little control over

More information

A Distributed Re-configurable Grid Workflow Engine

A Distributed Re-configurable Grid Workflow Engine A Distributed Re-configurable Grid Workflow Engine Jian Cao, Minglu Li, Wei Wei, and Shensheng Zhang Department of Computer Science & Technology, Shanghai Jiaotong University, 200030, Shanghai, P.R. China

More information

A RESOURCE MANAGEMENT FRAMEWORK FOR INTERACTIVE GRIDS

A RESOURCE MANAGEMENT FRAMEWORK FOR INTERACTIVE GRIDS A RESOURCE MANAGEMENT FRAMEWORK FOR INTERACTIVE GRIDS Raj Kumar, Vanish Talwar, Sujoy Basu Hewlett-Packard Labs 1501 Page Mill Road, MS 1181 Palo Alto, CA 94304 USA { raj.kumar,vanish.talwar,sujoy.basu}@hp.com

More information

Knowledge Discovery Services and Tools on Grids

Knowledge Discovery Services and Tools on Grids Knowledge Discovery Services and Tools on Grids DOMENICO TALIA DEIS University of Calabria ITALY talia@deis.unical.it Symposium ISMIS 2003, Maebashi City, Japan, Oct. 29, 2003 OUTLINE Introduction Grid

More information

Multi-path based Algorithms for Data Transfer in the Grid Environment

Multi-path based Algorithms for Data Transfer in the Grid Environment New Generation Computing, 28(2010)129-136 Ohmsha, Ltd. and Springer Multi-path based Algorithms for Data Transfer in the Grid Environment Muzhou XIONG 1,2, Dan CHEN 2,3, Hai JIN 1 and Song WU 1 1 School

More information

2-PHASE COMMIT PROTOCOL

2-PHASE COMMIT PROTOCOL 2-PHASE COMMIT PROTOCOL Jens Lechtenbörger, University of Münster, Germany SYNONYMS XA standard, distributed commit protocol DEFINITION The 2-phase commit (2PC) protocol is a distributed algorithm to ensure

More information

The Comparative Study of Machine Learning Algorithms in Text Data Classification*

The Comparative Study of Machine Learning Algorithms in Text Data Classification* The Comparative Study of Machine Learning Algorithms in Text Data Classification* Wang Xin School of Science, Beijing Information Science and Technology University Beijing, China Abstract Classification

More information

Lecture 9: MIMD Architectures

Lecture 9: MIMD Architectures Lecture 9: MIMD Architectures Introduction and classification Symmetric multiprocessors NUMA architecture Clusters Zebo Peng, IDA, LiTH 1 Introduction A set of general purpose processors is connected together.

More information

A priority based dynamic bandwidth scheduling in SDN networks 1

A priority based dynamic bandwidth scheduling in SDN networks 1 Acta Technica 62 No. 2A/2017, 445 454 c 2017 Institute of Thermomechanics CAS, v.v.i. A priority based dynamic bandwidth scheduling in SDN networks 1 Zun Wang 2 Abstract. In order to solve the problems

More information

COOCHING: Cooperative Prefetching Strategy for P2P Video-on-Demand System

COOCHING: Cooperative Prefetching Strategy for P2P Video-on-Demand System COOCHING: Cooperative Prefetching Strategy for P2P Video-on-Demand System Ubaid Abbasi and Toufik Ahmed CNRS abri ab. University of Bordeaux 1 351 Cours de la ibération, Talence Cedex 33405 France {abbasi,

More information

Dynamically Provisioning Distributed Systems to Meet Target Levels of Performance, Availability, and Data Quality

Dynamically Provisioning Distributed Systems to Meet Target Levels of Performance, Availability, and Data Quality Dynamically Provisioning Distributed Systems to Meet Target Levels of Performance, Availability, and Data Quality Amin Vahdat Department of Computer Science Duke University 1 Introduction Increasingly,

More information

Performance Improvement of Hardware-Based Packet Classification Algorithm

Performance Improvement of Hardware-Based Packet Classification Algorithm Performance Improvement of Hardware-Based Packet Classification Algorithm Yaw-Chung Chen 1, Pi-Chung Wang 2, Chun-Liang Lee 2, and Chia-Tai Chan 2 1 Department of Computer Science and Information Engineering,

More information

processes based on Message Passing Interface

processes based on Message Passing Interface Checkpointing and Migration of parallel processes based on Message Passing Interface Zhang Youhui, Wang Dongsheng, Zheng Weimin Department of Computer Science, Tsinghua University, China. Abstract This

More information

CAS 703 Software Design

CAS 703 Software Design Dr. Ridha Khedri Department of Computing and Software, McMaster University Canada L8S 4L7, Hamilton, Ontario Acknowledgments: Material based on Software by Tao et al. (Chapters 9 and 10) (SOA) 1 Interaction

More information