Two-Level Scheduling Technique for Mixed Best-Effort and QoS Job Arrays on Cluster Systems

Size: px
Start display at page:

Download "Two-Level Scheduling Technique for Mixed Best-Effort and QoS Job Arrays on Cluster Systems"

Transcription

1 Two-Level Scheduling Technique for Mixed Best-Effort and QoS Job Arrays on Cluster Systems Ekasit Kijsipongse, Suriya U-ruekolan, Sornthep Vannarat Large Scale Simulation Research Laboratory National Electronics and Computer Technology Center ABSTRACT We consider the problem of dynamic scheduling of mixed best-effort and QoS job arrays on cluster computers. Each job array consists of many independent jobs that arrive into the cluster over time. The information of job arrays, i.e. the number of jobs and the processing time of each job, are not known to the scheduler until they arrive. A user submits a job array in either the best-effort or QoS type. In case of the best-effort type, the scheduler accepts the job array and allocates approximately equal amount of computing resources to all users under fair sharing policy. For a QoS job array, which all jobs in the job array must be finished by the deadline, the scheduler performs the online admission control to immediately notify the user if the newly QoS job array is accepted or rejected. The scheduler must reorganize the existing schedule accordingly to ensure that the fairness of the best-effort and the deadlines of all accepted QoS job arrays are always guaranteed. We propose the online two-level scheduling technique to schedule best-effort and QoS job arrays on shared cluster computing resources to meet diverse requirements from users. Keywords: Job Scheduling, Cluster Computing, Admission Control. 1. INTRODUCTION The continuous expansion of user groups in High Performance Computing (HPC) has given a significant impact on the design of the job scheduling and management module, an essential part of an HPC system. The job scheduler must be able to response to the diversity of the user groups and their applications in finding an efficient resource allocation so that users can gain the most profit out of the HPC resources while being fair to other users. At present, there exists another type of applications called the parameter sweep applications which the applications are composed of a number of independent sequential jobs. Each job performs the same task on different parameters or subsets of data. Since the jobs are independent, there is no communication or dependencies among them, and thus makes this type of applications suitable to be executed on a loosely-coupled distributed HPC such as cluster computing system. These applications can often be found in data mining, simulation, computation biology or animation rendering; each of which may consist of thousands of independent jobs. In the context of job scheduler, relevant jobs in the parameter sweep applications are submitted in group as a job array. Therefore, the new job scheduler should support these parameter sweep applications in addition to the sequential and parallel jobs found in typical HPCs. The diversity of the user groups imposes a problem to the job scheduling and management on HPC as well since their requirements for the HPC are often different. In general, two main user groups can be identified at any HPC centers. The first group consists of users who need their jobs to be executed with QoS guarantee, a primary requirement from industrial users. In most cases, a QoS job is submitted with a given deadline. The scheduler must find ways to allocate HPC resources to complete the QoS jobs by the deadlines and this must be transparent to the users. The

2 immediate notification is also another required feature of a job scheduler. The faster the users know if the job will or will not be completed by its deadline, the better they can make further decision. The second group consists of users who use HPC for academic or testing purpose. They are provided with the best-effort policy such that their jobs will be executed if the computing resources are not occupied by the QoS jobs. The scheduler should prevent a particular best-effort user from dominating the HPC resources and it should allocate resources equally to all of them in this group. As there exist both QoS and best-effort jobs waiting for resources, the job scheduler must select an appropriate policy to allocate the computing resources to satisfy the requirements of all users. This paper presents the design and implementation of an two-level scheduler with admission control to support the resource allocation for aforementioned QoS and best-effort job arrays. It is a dynamic scheduler that works in the online mode which the arrival of jobs is not known in advance. The rest of the paper is organized as follows. Section 2 gives an overview of relevant work in this area. Section 3 presents the design and implementation of the two-level scheduler with admission control. Section 4 draws conclusions and discusses the future work. In this paper, the terms job and job array will be used interchangeably unless they needs to be specific. 2. RELATED WORKS Online deadline scheduling on a single machine has been introduced in [1] where they emphasized on the immediate notification ability of job admission control. Similarly, scheduling jobs with equal processing time on multiple identical machines is discussed in [2, 3, 4]. Kim and Chwa [5] considered the online deadline scheduling on multiple resources but without admission control, which makes the job able to be rejected anytime before its deadline. The study of job array scheduling also appears in the context of scheduling Bag-of-Task (BoT) applications [6, 7, 8]. However, they dedicated to the BoT scheduling to meet a single objective such as minimizing makespan or minimizing the maximum completion time rather than to meet distinct requirements for different user groups as in our work. Anderson et al. [9] studied the offline scheduling for job arrays in animation rendering to maximize the utility value and meet their deadlines. The problem of job scheduling with mixed workload containing a steady incoming stream of best-effort and real-time jobs on clusters is presented in [10]. Libra [11] is a job scheduler on cluster systems that can provide immediate notification to users but it requires that at least a node is available to execute the job immediately. SGE [12], Condor [13], and Maui [14] are open source schedulers that support running job arrays on clusters or network of work stations; yet the admission control with immediate notification has not been implemented. The commercial Moab scheduler [15] provides a limited support for deadline scheduling 3. DESIGN AND IMPLEMENTATION OF THE TWO-LEVEL SCHEDULER Cluster computing systems typically consist of one frontend node and a number of compute nodes. Users submit jobs into the frontend node where a job scheduler interacts with the cluster resource manager to distribute jobs to compute nodes for execution. We use Torque [16] for the resource manager to provide the primitive job manipulation commands such as start, hold, cancel, and monitor jobs. Torque is an open source variant of PBS batch system and it has been widely used in many HPC centers over years. The decision on which job is started on which compute node is made by an external scheduler (Torque is bundled with a built-in but simple scheduler pbs_sched). Torque consists of two main components, pbs_server and pbs_mom, as illustrated in Figure 1. The pbs_server is a central server running on the frontend node to receive new jobs into job queues. It communicates with pbs_mom to collect the status of all compute nodes and jobs. The pbs_mom runs on each compute node to monitor status of the compute node and reports it to the pbs_server. When the pbs_server needs job scheduling, it sends a trigger message to an external scheduler. Trigger messages are sent according to several

3 events such as new jobs arrive, executing jobs terminate, or scheduling time is reached. After receiving a trigger message, the external scheduler module interacts with the pbs_server to perform job scheduling. We implement our two-level scheduler on PluS scheduler [17], a Java-based scheduler that can communicate with Torque. The two-level scheduler acts as a matchmaker between jobs and computing resources. When computing resources become available, the scheduler selects the jobs to be executed in the resources to meet our objective: the QoS jobs are always guaranteed to be completed within the deadlines while the best-effort jobs equally share the computing resources. To decide if the QoS jobs are accepted or rejected, we rely on the admission control module. For a QoS Job, the modified qsub command takes additional two arguments, -l deadline=[[[[cc]yy]mm]dd]hhmm[.ss] and -l walltime=hh:mm:ss for the deadline and the processing time of the submitted job, respectively. The qsub command first contacts the admission control for the immediate notification of acceptance. The job is rejected if resources are not enough for the job and all previously accepted QoS jobs to finish by their deadlines; otherwise the job is accepted and waits in the job queue for scheduling later. For a best-effort job, the qsub command works as usual, i.e. the job is put into the job queue. We assume in this work that the system is a homogeneous cluster, or in other words, all compute nodes are identical. Figure 1. Two-level scheduler with Torque resource manager. 3.1 ONLINE ADMISSION CONTROL The online admission control needs to ensure that the new QoS job can be completed by its deadline without violating the deadline of previously accepted jobs. The feasibility testing algorithm to accept a new job is given in Listing 1. The available time f(m) of the machine m is the earliest time that the machine is free to be used. When a job is dispatched to execute on a machine, its available time is updated accordingly. The p(j) and d(j) are the processing time and the deadline of job j given by the user, respectively. To test an incoming job arrays, every job in it

4 and all previously job arrays has to be checked if it can be completed by the deadline. We use the Earliest Deadline First (EDF) algorithm to select the job to be tested. If all QoS jobs can pass the test, the incoming job array is then admitted into the job queue of Torque. Algorithm IsFeasible Input : The incoming job arrays Output : True or False Listing 1. Feasible testing algorithm. let E be the list of previously accepted and the incoming QoS job arrays sorted by their deadlines for each job array A E do for each job j job array A do let m be the machine with the minimum available time f(m) if f(m) + p(j) > d(j) then return false else f(m) = f(m) + p(j) end if return true 3.2 FIRST LEVEL: DYNAMIC QoS SCHEDULING The jobs in the job queue are mixed with QoS and best-effort jobs. As QoS jobs takes higher priority than best-effort jobs, all cluster resources are always allocated to the QoS jobs first if possible. Thus, when there is an available compute node, we apply the EDF algorithm to select the next job to be dispatched so that the job with the earliest deadline always starts first. Though using the EDF algorithm can meet our requirement, it is a research challenge to see if other algorithms can be more efficient. If there is no QoS job existed in the job queue, fairshare scheduling is invoked for the best-effort jobs as illustrated in Figure 2. Figure 2. QoS and fairshare schedulers 3.3 SECOND LEVEL: DYNAMIC FAIRSHARE SCHEDULING Fairshare scheduling allows users to use cluster resources equally. If user A has done twice as much work as user B over a period of time, then user B should thereafter be able to do twice as much work as user A. The fairshare scheduler records the CPU usage of each user for his besteffort jobs executed so far and selects the next job belonging to the user who has the smallest CPU usage. The time in fairshare scheduling is discretized into a number of time frames. At the end of each time frame, the CPU usage of each individual job in a job array is calculated from the resources_used.cput field of Torque s job status which represents the CPU time used by this job since started. To limit the impact of usage data which are too old, the CPU usages are

5 exponentially decayed by time so that higher weight is put on more recent data when calculating the accumulative CPU usages over multiple time frames. The overall algorithm of the fairshare scheduling is shown in Listing 2. The CPU usage history is denoted by cpuusage, the set of running jobs at the previous time frame by R t-1, and the set of running jobs at the current time frame by R t. The used CPU times of each job are stored in cput t and cput t-1 for the current and previous time frame, respectively. The parameter FSDECAY is the exponential decay factor, and FSINTERVAL is the duration of each time frame. The values of these parameters can be set to suit each specific need. For more sophisticated fairshare scheduling such as group or project based fairshare, readers may refer to [18] Listing 2. Fairshare scheduling algorithm. Algorithm FairshareScheduling Input : A set of best-effort job arrays J Output : The selected job k for each user u do cpuusage{u} = FSDECAY * cpuusage{u} for each running job j R t-1 and j is a best-effort job do if j R t continue let u be the owner of job j cpuusage{u} = cpuusage{u} + ( cput t(j) - cput t-1(j) ) let A be the best-effort job array ( J) of the user with the smallest cpuusage let k be an arbitrary job in job array A return k 4. CONCLUSIONS We have proposed the two-level scheduling with admission control to support resource allocation in cluster computing systems for mixed QoS and best-effort job arrays. The implementation is based on the combination between two well-known job scheduling algorithms. Job admission control and the dynamic scheduling of QoS job arrays are accomplished by the EDF algorithm to ensure the completion time by their deadlines. The fairshare algorithm is used in scheduling of best-effort job arrays so that the computing resources are equally occupied by all users. The proposed scheduler can enrich the HPC systems to support more diverse user requirements in this complex world. To the users, the system is more responsive and versatile. To the resource owners, the new scheduler allows the resources to be used in more efficient manner. Furthermore, the scheduler can interface with many resource managers such as Torque, SGE or condor. In the future work, we plan to extend our scheduler for cost-based admission control and add a global resource sharing policy between the entire QoS and best-effort user groups. REFERENCES 1. M. H. Goldwasser and B. Kerbikov, Admission control with immediate notification, J. of Scheduling, vol. 6, no. 3, pp , J. Ding and G. Zhang, Online scheduling with hard deadlines on parallel machines, in Algorithmic Aspects in Information and Management, Second International Conference, Proceedings, Lecture Notes in Computer Science, S.-W. Cheng and C. K. Poon, Eds., vol Springer, pp , 2006.

6 3. J. Ding, T. Ebenlendr, J. Sgall, and G. Zhang, Online scheduling of equal-length jobs on parallel machines, in Algorithms - ESA 2007, 15th Annual European Symposium, Proceedings, Lecture Notes in Computer Science, vol Springer, pp , M. H. Goldwasser and M. Pedigo, Online nonpreemptive scheduling of equal-length jobs on two identical machines, ACM Trans. Algorithms, vol. 5, no. 1, pp. 1-18, J.-H. Kim and K.-Y. Chwa, On-line deadline scheduling on multiple resources, in COCOON '01: Proceedings of the 7th Annual International Conference on Computing and Combinatorics. Springer-Verlag, pp , W. Cirne, D. Paranhos, L. Costa, E. Santos-Neto, F. Brasileiro, J. Sauve, F. A. B. Silva, C. O. Barros, C. Silveira, and C. Silveira, Running bag-of-tasks applications on computational grids: the mygrid approach, in Parallel Processing, Proceedings International Conference on, pp , Y. C. Lee and A. Y. Zomaya, Practical scheduling of bag-of-tasks applications on grids with dynamic resilience, IEEE Trans. Comput., vol. 56, no. 6, pp , A. Benoit, L. Marchal, J.-F. Pineau, Y. Robert, and F. Vivien, Offline and online masterworker scheduling of concurrent bags-of-tasks on heterogeneous platforms, in 22nd IEEE International Symposium on Parallel and Distributed Processing, IPDPS 2008, pp. 1-8, E. Anderson, D. Beyer, K. Chaudhuri, T. Kelly, N. Salazar, C. Santos, R. Swaminathan, R. Tarjan, J. Wiener, and Y. Zhou, Value-maximizing deadline scheduling and its application to animation rendering, in SPAA '05: Proceedings of the seventeenth annual ACM symposium on Parallelism in algorithms and architectures. ACM, pp , Y. Zhang and A. Sivasubramaniam, Scheduling best-effort and real-time pipelined applications on time-shared clusters, in SPAA '01: Proceedings of the thirteenth annual ACM symposium on Parallel algorithms and architectures. ACM, pp , J. Sherwani, N. Ali, N. Lotia, Z. Hayat, and R. Buyya, Libra: a computational economybased job scheduling system for clusters, Softw. Pract. Exper., vol. 34, no. 6, pp , Sun grid engine, Condor project, Maui scheduler, Moab workload manager, Torque resource manager, H. Nakada, A. Takefusa, K. Ookubo, M. Kishimoto, T. Kudoh, Y. Tanaka, and S. Sekiguchi, Design and implementation of a local scheduling system with advance reservation for coallocation on the grid, in CIT '06: Proceedings of the Sixth IEEE International Conference on Computer and Information Technology. p. 65, D. B. Jackson, Q. Snell, and M. J. Clement, Core algorithms of the maui scheduler, in JSSPP '01: Revised Papers from the 7th International Workshop on Job Scheduling Strategies for Parallel Processing. Springer-Verlag, pp , 2001.

LSKA 2010 Survey Report Job Scheduler

LSKA 2010 Survey Report Job Scheduler LSKA 2010 Survey Report Job Scheduler Graduate Institute of Communication Engineering {r98942067, r98942112}@ntu.edu.tw March 31, 2010 1. Motivation Recently, the computing becomes much more complex. However,

More information

159.735. Final Report. Cluster Scheduling. Submitted by: Priti Lohani 04244354

159.735. Final Report. Cluster Scheduling. Submitted by: Priti Lohani 04244354 159.735 Final Report Cluster Scheduling Submitted by: Priti Lohani 04244354 1 Table of contents: 159.735... 1 Final Report... 1 Cluster Scheduling... 1 Table of contents:... 2 1. Introduction:... 3 1.1

More information

Scheduling Algorithms for Dynamic Workload

Scheduling Algorithms for Dynamic Workload Managed by Scheduling Algorithms for Dynamic Workload Dalibor Klusáček (MU) Hana Rudová (MU) Ranieri Baraglia (CNR - ISTI) Gabriele Capannini (CNR - ISTI) Marco Pasquali (CNR ISTI) Outline Motivation &

More information

BACKFILLING STRATEGIES FOR SCHEDULING STREAMS OF JOBS ON COMPUTATIONAL FARMS

BACKFILLING STRATEGIES FOR SCHEDULING STREAMS OF JOBS ON COMPUTATIONAL FARMS BACKFILLING STRATEGIES FOR SCHEDULING STREAMS OF JOBS ON COMPUTATIONAL FARMS A.D.Techiouba, G.Capannini, Ranieri Baraglia, D.Puppin, M.Pasquali ISTI,CNR Via Moruzzi, 1 Pisa, Italy techioub@cli.di.unipi.it

More information

A Bi-Objective Approach for Cloud Computing Systems

A Bi-Objective Approach for Cloud Computing Systems A Bi-Objective Approach for Cloud Computing Systems N.Geethanjali 1, M.Ramya 2 Assistant Professor, Department of Computer Science, Christ The King Engineering College 1, 2 ABSTRACT: There are Various

More information

GC3: Grid Computing Competence Center Cluster computing, I Batch-queueing systems

GC3: Grid Computing Competence Center Cluster computing, I Batch-queueing systems GC3: Grid Computing Competence Center Cluster computing, I Batch-queueing systems Riccardo Murri, Sergio Maffioletti Grid Computing Competence Center, Organisch-Chemisches Institut, University of Zurich

More information

Batch Systems. provide a mechanism for submitting, launching, and tracking jobs on a shared resource

Batch Systems. provide a mechanism for submitting, launching, and tracking jobs on a shared resource PBS INTERNALS PBS & TORQUE PBS (Portable Batch System)-software system for managing system resources on workstations, SMP systems, MPPs and vector computers. It was based on Network Queuing System (NQS)

More information

Heterogeneous Workload Consolidation for Efficient Management of Data Centers in Cloud Computing

Heterogeneous Workload Consolidation for Efficient Management of Data Centers in Cloud Computing Heterogeneous Workload Consolidation for Efficient Management of Data Centers in Cloud Computing Deep Mann ME (Software Engineering) Computer Science and Engineering Department Thapar University Patiala-147004

More information

A Multi-criteria Class-based Job Scheduler for Large Computing Farms

A Multi-criteria Class-based Job Scheduler for Large Computing Farms A Multi-criteria Class-based Job Scheduler for Large Computing Farms R. Baraglia 1, P. Dazzi 1, and R. Ferrini 1 1 ISTI A. Faedo, CNR, Pisa, Italy Abstract In this paper we propose a new multi-criteria

More information

How To Balance A Web Server With Remaining Capacity

How To Balance A Web Server With Remaining Capacity Remaining Capacity Based Load Balancing Architecture for Heterogeneous Web Server System Tsang-Long Pao Dept. Computer Science and Engineering Tatung University Taipei, ROC Jian-Bo Chen Dept. Computer

More information

On the Amplitude of the Elasticity Offered by Public Cloud Computing Providers

On the Amplitude of the Elasticity Offered by Public Cloud Computing Providers On the Amplitude of the Elasticity Offered by Public Cloud Computing Providers Rostand Costa a,b, Francisco Brasileiro a a Federal University of Campina Grande Systems and Computing Department, Distributed

More information

A Design of Resource Fault Handling Mechanism using Dynamic Resource Reallocation for the Resource and Job Management System

A Design of Resource Fault Handling Mechanism using Dynamic Resource Reallocation for the Resource and Job Management System A Design of Resource Fault Handling Mechanism using Dynamic Resource Reallocation for the Resource and Job Management System Young-Ho Kim, Eun-Ji Lim, Gyu-Il Cha, Seung-Jo Bae Electronics and Telecommunications

More information

Grid Computing Approach for Dynamic Load Balancing

Grid Computing Approach for Dynamic Load Balancing International Journal of Computer Sciences and Engineering Open Access Review Paper Volume-4, Issue-1 E-ISSN: 2347-2693 Grid Computing Approach for Dynamic Load Balancing Kapil B. Morey 1*, Sachin B. Jadhav

More information

Resource Allocation Schemes for Gang Scheduling

Resource Allocation Schemes for Gang Scheduling Resource Allocation Schemes for Gang Scheduling B. B. Zhou School of Computing and Mathematics Deakin University Geelong, VIC 327, Australia D. Walsh R. P. Brent Department of Computer Science Australian

More information

PERFORMANCE ANALYSIS OF PaaS CLOUD COMPUTING SYSTEM

PERFORMANCE ANALYSIS OF PaaS CLOUD COMPUTING SYSTEM PERFORMANCE ANALYSIS OF PaaS CLOUD COMPUTING SYSTEM Akmal Basha 1 Krishna Sagar 2 1 PG Student,Department of Computer Science and Engineering, Madanapalle Institute of Technology & Science, India. 2 Associate

More information

A two-level scheduler to dynamically schedule a stream of batch jobs in large-scale grids

A two-level scheduler to dynamically schedule a stream of batch jobs in large-scale grids Managed by A two-level scheduler to dynamically schedule a stream of batch jobs in large-scale grids M. Pasquali, R. Baraglia, G. Capannini, L. Ricci, and D. Laforenza 7th Meeting of the Institute on Resource

More information

ISSN: 2321-7782 (Online) Volume 2, Issue 5, May 2014 International Journal of Advance Research in Computer Science and Management Studies

ISSN: 2321-7782 (Online) Volume 2, Issue 5, May 2014 International Journal of Advance Research in Computer Science and Management Studies ISSN: 2321-7782 (Online) Volume 2, Issue 5, May 2014 International Journal of Advance Research in Computer Science and Management Studies Research Article / Survey Paper / Case Study Available online at:

More information

How To Use A Cloud For A Local Cluster

How To Use A Cloud For A Local Cluster Marcos Dias de Assunção 1,2, Alexandre di Costanzo 1 and Rajkumar Buyya 1 1 Department of Computer Science and Software Engineering 2 National ICT Australia (NICTA) Victoria Research Laboratory The University

More information

Cost-effective Provisioning and Scheduling of Deadline-constrained Applications in Hybrid Clouds

Cost-effective Provisioning and Scheduling of Deadline-constrained Applications in Hybrid Clouds Cost-effective Provisioning and Scheduling of Deadline-constrained Applications in Hybrid Clouds Rodrigo N. Calheiros and Rajkumar Buyya Cloud Computing and Distributed Systems (CLOUDS) Laboratory Department

More information

Grid Scheduling Dictionary of Terms and Keywords

Grid Scheduling Dictionary of Terms and Keywords Grid Scheduling Dictionary Working Group M. Roehrig, Sandia National Laboratories W. Ziegler, Fraunhofer-Institute for Algorithms and Scientific Computing Document: Category: Informational June 2002 Status

More information

Reverse Auction-based Resource Allocation Policy for Service Broker in Hybrid Cloud Environment

Reverse Auction-based Resource Allocation Policy for Service Broker in Hybrid Cloud Environment Reverse Auction-based Resource Allocation Policy for Service Broker in Hybrid Cloud Environment Sunghwan Moon, Jaekwon Kim, Taeyoung Kim, Jongsik Lee Department of Computer and Information Engineering,

More information

Map-Parallel Scheduling (mps) using Hadoop environment for job scheduler and time span for Multicore Processors

Map-Parallel Scheduling (mps) using Hadoop environment for job scheduler and time span for Multicore Processors Map-Parallel Scheduling (mps) using Hadoop environment for job scheduler and time span for Sudarsanam P Abstract G. Singaravel Parallel computing is an base mechanism for data process with scheduling task,

More information

Provisioning and Resource Management at Large Scale (Kadeploy and OAR)

Provisioning and Resource Management at Large Scale (Kadeploy and OAR) Provisioning and Resource Management at Large Scale (Kadeploy and OAR) Olivier Richard Laboratoire d Informatique de Grenoble (LIG) Projet INRIA Mescal 31 octobre 2007 Olivier Richard ( Laboratoire d Informatique

More information

Fair Scheduling Algorithm with Dynamic Load Balancing Using In Grid Computing

Fair Scheduling Algorithm with Dynamic Load Balancing Using In Grid Computing Research Inventy: International Journal Of Engineering And Science Vol.2, Issue 10 (April 2013), Pp 53-57 Issn(e): 2278-4721, Issn(p):2319-6483, Www.Researchinventy.Com Fair Scheduling Algorithm with Dynamic

More information

Scheduling Allowance Adaptability in Load Balancing technique for Distributed Systems

Scheduling Allowance Adaptability in Load Balancing technique for Distributed Systems Scheduling Allowance Adaptability in Load Balancing technique for Distributed Systems G.Rajina #1, P.Nagaraju #2 #1 M.Tech, Computer Science Engineering, TallaPadmavathi Engineering College, Warangal,

More information

An Evaluation of Economy-based Resource Trading and Scheduling on Computational Power Grids for Parameter Sweep Applications

An Evaluation of Economy-based Resource Trading and Scheduling on Computational Power Grids for Parameter Sweep Applications An Evaluation of Economy-based Resource Trading and Scheduling on Computational Power Grids for Parameter Sweep Applications Rajkumar Buyya, Jonathan Giddy, and David Abramson School of Computer Science

More information

Libra: a computational economy-based job scheduling system for clusters

Libra: a computational economy-based job scheduling system for clusters SOFTWARE PRACTICE AND EXPERIENCE Softw. Pract. Exper. 2004; 34:573 590 (DOI: 10.1002/spe.581) Libra: a computational economy-based job scheduling system for clusters Jahanzeb Sherwani 1, Nosheen Ali 1,

More information

The Moab Scheduler. Dan Mazur, McGill HPC daniel.mazur@mcgill.ca Aug 23, 2013

The Moab Scheduler. Dan Mazur, McGill HPC daniel.mazur@mcgill.ca Aug 23, 2013 The Moab Scheduler Dan Mazur, McGill HPC daniel.mazur@mcgill.ca Aug 23, 2013 1 Outline Fair Resource Sharing Fairness Priority Maximizing resource usage MAXPS fairness policy Minimizing queue times Should

More information

Decentralized Load Balancing for Heterogeneous Grids

Decentralized Load Balancing for Heterogeneous Grids Decentralized Load Balancing for Heterogeneous Grids Issam Al-Azzoni and Douglas G. Down Department of Computing and Software McMaster University Hamilton, Ontario, Canada alazzoi@mcmaster.ca, downd@mcmaster.ca

More information

Efficient Scheduling Of On-line Services in Cloud Computing Based on Task Migration

Efficient Scheduling Of On-line Services in Cloud Computing Based on Task Migration Efficient Scheduling Of On-line Services in Cloud Computing Based on Task Migration 1 Harish H G, 2 Dr. R Girisha 1 PG Student, 2 Professor, Department of CSE, PESCE Mandya (An Autonomous Institution under

More information

COMPARISON OF MULTI-CRITERIA SCHEDULING TECHNIQUES

COMPARISON OF MULTI-CRITERIA SCHEDULING TECHNIQUES COMPARISON OF MULTI-CRITERIA SCHEDULING TECHNIQUES Dalibor Klusáček and Hana Rudová Faculty of Informatics, Masaryk University Botanická 68a, Brno, Czech Republic {xklusac, hanka}@fi.muni.cz Ranieri Baraglia

More information

Batch Scheduling and Resource Management

Batch Scheduling and Resource Management Batch Scheduling and Resource Management Luke Tierney Department of Statistics & Actuarial Science University of Iowa October 18, 2007 Luke Tierney (U. of Iowa) Batch Scheduling and Resource Management

More information

Profit Maximization and Power Management of Green Data Centers Supporting Multiple SLAs

Profit Maximization and Power Management of Green Data Centers Supporting Multiple SLAs Profit Maximization and Power Management of Green Data Centers Supporting Multiple SLAs Mahdi Ghamkhari and Hamed Mohsenian-Rad Department of Electrical Engineering University of California at Riverside,

More information

Real Time Network Server Monitoring using Smartphone with Dynamic Load Balancing

Real Time Network Server Monitoring using Smartphone with Dynamic Load Balancing www.ijcsi.org 227 Real Time Network Server Monitoring using Smartphone with Dynamic Load Balancing Dhuha Basheer Abdullah 1, Zeena Abdulgafar Thanoon 2, 1 Computer Science Department, Mosul University,

More information

A General Distributed Scalable Peer to Peer Scheduler for Mixed Tasks in Grids

A General Distributed Scalable Peer to Peer Scheduler for Mixed Tasks in Grids A General Distributed Scalable Peer to Peer Scheduler for Mixed Tasks in Grids Cong Liu, Sanjeev Baskiyar *, and Shuang Li Dept. of Computer Science and Software Engineering Auburn University, Auburn,

More information

Load Balancing in Fault Tolerant Video Server

Load Balancing in Fault Tolerant Video Server Load Balancing in Fault Tolerant Video Server # D. N. Sujatha*, Girish K*, Rashmi B*, Venugopal K. R*, L. M. Patnaik** *Department of Computer Science and Engineering University Visvesvaraya College of

More information

Load Balancing Scheduling with Shortest Load First

Load Balancing Scheduling with Shortest Load First , pp. 171-178 http://dx.doi.org/10.14257/ijgdc.2015.8.4.17 Load Balancing Scheduling with Shortest Load First Ranjan Kumar Mondal 1, Enakshmi Nandi 2 and Debabrata Sarddar 3 1 Department of Computer Science

More information

A High Performance Computing Scheduling and Resource Management Primer

A High Performance Computing Scheduling and Resource Management Primer LLNL-TR-652476 A High Performance Computing Scheduling and Resource Management Primer D. H. Ahn, J. E. Garlick, M. A. Grondona, D. A. Lipari, R. R. Springmeyer March 31, 2014 Disclaimer This document was

More information

Generic Grid Computing Tools for Resource and Project Management

Generic Grid Computing Tools for Resource and Project Management Generic Grid Computing Tools for and Project Management Erik Elmroth Dept. of Computing Science & HPC2N Umeå University, Sweden Overall objectives Short term: Generic infrastructure components for resource

More information

A CP Scheduler for High-Performance Computers

A CP Scheduler for High-Performance Computers A CP Scheduler for High-Performance Computers Thomas Bridi, Michele Lombardi, Andrea Bartolini, Luca Benini, and Michela Milano {thomas.bridi,michele.lombardi2,a.bartolini,luca.benini,michela.milano}@

More information

5 Performance Management for Web Services. Rolf Stadler School of Electrical Engineering KTH Royal Institute of Technology. stadler@ee.kth.

5 Performance Management for Web Services. Rolf Stadler School of Electrical Engineering KTH Royal Institute of Technology. stadler@ee.kth. 5 Performance Management for Web Services Rolf Stadler School of Electrical Engineering KTH Royal Institute of Technology stadler@ee.kth.se April 2008 Overview Service Management Performance Mgt QoS Mgt

More information

Optimization for QoS on Web-Service-Based Systems with Tasks Deadlines 1

Optimization for QoS on Web-Service-Based Systems with Tasks Deadlines 1 Optimization for QoS on Web-Service-Based Systems with Tasks Deadlines 1 Luís Fernando Orleans Departamento de Engenharia Informática Universidade de Coimbra Coimbra, Portugal lorleans@dei.uc.pt Pedro

More information

MEASURING PERFORMANCE OF DYNAMIC LOAD BALANCING ALGORITHMS IN DISTRIBUTED COMPUTING APPLICATIONS

MEASURING PERFORMANCE OF DYNAMIC LOAD BALANCING ALGORITHMS IN DISTRIBUTED COMPUTING APPLICATIONS MEASURING PERFORMANCE OF DYNAMIC LOAD BALANCING ALGORITHMS IN DISTRIBUTED COMPUTING APPLICATIONS Priyesh Kanungo 1 Professor and Senior Systems Engineer (Computer Centre), School of Computer Science and

More information

Resource Management and Job Scheduling

Resource Management and Job Scheduling Resource Management and Job Scheduling Jenett Tillotson Senior Cluster System Administrator Indiana University May 18 18-22 May 2015 1 Resource Managers Keep track of resources Nodes: CPUs, disk, memory,

More information

FAULT TOLERANCE FOR MULTIPROCESSOR SYSTEMS VIA TIME REDUNDANT TASK SCHEDULING

FAULT TOLERANCE FOR MULTIPROCESSOR SYSTEMS VIA TIME REDUNDANT TASK SCHEDULING FAULT TOLERANCE FOR MULTIPROCESSOR SYSTEMS VIA TIME REDUNDANT TASK SCHEDULING Hussain Al-Asaad and Alireza Sarvi Department of Electrical & Computer Engineering University of California Davis, CA, U.S.A.

More information

Scheduling Video Stream Transmissions for Distributed Playback over Mobile Cellular Networks

Scheduling Video Stream Transmissions for Distributed Playback over Mobile Cellular Networks Scheduling Video Stream Transmissions for Distributed Playback over Mobile Cellular Networks Kam-Yiu Lam 1, Joe Yuen 1, Sang H. Son 2 and Edward Chan 1 Department of Computer Science 1 City University

More information

OPERATING SYSTEMS SCHEDULING

OPERATING SYSTEMS SCHEDULING OPERATING SYSTEMS SCHEDULING Jerry Breecher 5: CPU- 1 CPU What Is In This Chapter? This chapter is about how to get a process attached to a processor. It centers around efficient algorithms that perform

More information

HYBRID ACO-IWD OPTIMIZATION ALGORITHM FOR MINIMIZING WEIGHTED FLOWTIME IN CLOUD-BASED PARAMETER SWEEP EXPERIMENTS

HYBRID ACO-IWD OPTIMIZATION ALGORITHM FOR MINIMIZING WEIGHTED FLOWTIME IN CLOUD-BASED PARAMETER SWEEP EXPERIMENTS HYBRID ACO-IWD OPTIMIZATION ALGORITHM FOR MINIMIZING WEIGHTED FLOWTIME IN CLOUD-BASED PARAMETER SWEEP EXPERIMENTS R. Angel Preethima 1, Margret Johnson 2 1 Student, Computer Science and Engineering, Karunya

More information

A REVIEW PAPER ON THE HADOOP DISTRIBUTED FILE SYSTEM

A REVIEW PAPER ON THE HADOOP DISTRIBUTED FILE SYSTEM A REVIEW PAPER ON THE HADOOP DISTRIBUTED FILE SYSTEM Sneha D.Borkar 1, Prof.Chaitali S.Surtakar 2 Student of B.E., Information Technology, J.D.I.E.T, sborkar95@gmail.com Assistant Professor, Information

More information

Batch Job Analysis to Improve the Success Rate in HPC

Batch Job Analysis to Improve the Success Rate in HPC Batch Job Analysis to Improve the Success Rate in HPC 1 JunWeon Yoon, 2 TaeYoung Hong, 3 ChanYeol Park, 4 HeonChang Yu 1, First Author KISTI and Korea University, jwyoon@kisti.re.kr 2,3, KISTI,tyhong@kisti.re.kr,chan@kisti.re.kr

More information

Manjrasoft Market Oriented Cloud Computing Platform

Manjrasoft Market Oriented Cloud Computing Platform Manjrasoft Market Oriented Cloud Computing Platform Innovative Solutions for 3D Rendering Aneka is a market oriented Cloud development and management platform with rapid application development and workload

More information

Enabling Interactive Jobs in Virtualized Data Centers (Extended Abstract)

Enabling Interactive Jobs in Virtualized Data Centers (Extended Abstract) 1 Enabling Interactive Jobs in Virtualized Data Centers (Extended Abstract) John Paul Walters, Bhagyashree Bantwal, and Vipin Chaudhary Department of Computer Science and Engineering University at Buffalo,

More information

Optimizing Shared Resource Contention in HPC Clusters

Optimizing Shared Resource Contention in HPC Clusters Optimizing Shared Resource Contention in HPC Clusters Sergey Blagodurov Simon Fraser University Alexandra Fedorova Simon Fraser University Abstract Contention for shared resources in HPC clusters occurs

More information

Web-based Dynamic Scheduling Platform for Grid Computing

Web-based Dynamic Scheduling Platform for Grid Computing IJCSNS International Journal of Computer Science and Network Security, VOL.6 No.5B, May 2006 67 Web-based Dynamic Scheduling Platform for Grid Computing Oh-han Kang, and Sang-seong Kang, Dept. of Computer

More information

Optimal Service Pricing for a Cloud Cache

Optimal Service Pricing for a Cloud Cache Optimal Service Pricing for a Cloud Cache K.SRAVANTHI Department of Computer Science & Engineering (M.Tech.) Sindura College of Engineering and Technology Ramagundam,Telangana G.LAKSHMI Asst. Professor,

More information

Developing Scalable Smart Grid Infrastructure to Enable Secure Transmission System Control

Developing Scalable Smart Grid Infrastructure to Enable Secure Transmission System Control Developing Scalable Smart Grid Infrastructure to Enable Secure Transmission System Control EP/K006487/1 UK PI: Prof Gareth Taylor (BU) China PI: Prof Yong-Hua Song (THU) Consortium UK Members: Brunel University

More information

SCHEDULING IN CLOUD COMPUTING

SCHEDULING IN CLOUD COMPUTING SCHEDULING IN CLOUD COMPUTING Lipsa Tripathy, Rasmi Ranjan Patra CSA,CPGS,OUAT,Bhubaneswar,Odisha Abstract Cloud computing is an emerging technology. It process huge amount of data so scheduling mechanism

More information

Resource Allocation Avoiding SLA Violations in Cloud Framework for SaaS

Resource Allocation Avoiding SLA Violations in Cloud Framework for SaaS Resource Allocation Avoiding SLA Violations in Cloud Framework for SaaS Shantanu Sasane Abhilash Bari Kaustubh Memane Aniket Pathak Prof. A. A.Deshmukh University of Pune University of Pune University

More information

Lecture Outline Overview of real-time scheduling algorithms Outline relative strengths, weaknesses

Lecture Outline Overview of real-time scheduling algorithms Outline relative strengths, weaknesses Overview of Real-Time Scheduling Embedded Real-Time Software Lecture 3 Lecture Outline Overview of real-time scheduling algorithms Clock-driven Weighted round-robin Priority-driven Dynamic vs. static Deadline

More information

Titolo del paragrafo. Titolo del documento - Sottotitolo documento The Benefits of Pushing Real-Time Market Data via a Web Infrastructure

Titolo del paragrafo. Titolo del documento - Sottotitolo documento The Benefits of Pushing Real-Time Market Data via a Web Infrastructure 1 Alessandro Alinone Agenda Introduction Push Technology: definition, typology, history, early failures Lightstreamer: 3rd Generation architecture, true-push Client-side push technology (Browser client,

More information

New Issues and New Capabilities in HPC Scheduling with the Maui Scheduler

New Issues and New Capabilities in HPC Scheduling with the Maui Scheduler New Issues and New Capabilities in HPC Scheduling with the Maui Scheduler I.Introduction David B Jackson Center for High Performance Computing, University of Utah Much has changed in a few short years.

More information

Dynamic Scheduling and Pricing in Wireless Cloud Computing

Dynamic Scheduling and Pricing in Wireless Cloud Computing Dynamic Scheduling and Pricing in Wireless Cloud Computing R.Saranya 1, G.Indra 2, Kalaivani.A 3 Assistant Professor, Dept. of CSE., R.M.K.College of Engineering and Technology, Puduvoyal, Chennai, India.

More information

A Queueing Theory Approach to Pareto Optimal Bags-of-Tasks Scheduling on Clouds

A Queueing Theory Approach to Pareto Optimal Bags-of-Tasks Scheduling on Clouds A Queueing Theory Approach to Pareto Optimal Bags-of-Tasks Scheduling on Clouds Cosmin Dumitru 1, Ana-Maria Oprescu 1, Miroslav Živković1, Rob van der Mei 2, Paola Grosso 1, and Cees de Laat 1 1 System

More information

Figure 1. The cloud scales: Amazon EC2 growth [2].

Figure 1. The cloud scales: Amazon EC2 growth [2]. - Chung-Cheng Li and Kuochen Wang Department of Computer Science National Chiao Tung University Hsinchu, Taiwan 300 shinji10343@hotmail.com, kwang@cs.nctu.edu.tw Abstract One of the most important issues

More information

GENETIC-BASED SOLUTIONS FOR INDEPENDENT BATCH SCHEDULING IN DATA GRIDS

GENETIC-BASED SOLUTIONS FOR INDEPENDENT BATCH SCHEDULING IN DATA GRIDS GENETIC-BASED SOLUTIONS FOR INDEPENDENT BATCH SCHEDULING IN DATA GRIDS Joanna Ko lodziej Cracow University of Technology, Poland Email: jokolodziej@pk.edu.pl Samee U. Khan North Dakota State University

More information

Efficient Data Replication Scheme based on Hadoop Distributed File System

Efficient Data Replication Scheme based on Hadoop Distributed File System , pp. 177-186 http://dx.doi.org/10.14257/ijseia.2015.9.12.16 Efficient Data Replication Scheme based on Hadoop Distributed File System Jungha Lee 1, Jaehwa Chung 2 and Daewon Lee 3* 1 Division of Supercomputing,

More information

A Distributed Render Farm System for Animation Production

A Distributed Render Farm System for Animation Production A Distributed Render Farm System for Animation Production Jiali Yao, Zhigeng Pan *, Hongxin Zhang State Key Lab of CAD&CG, Zhejiang University, Hangzhou, 310058, China {yaojiali, zgpan, zhx}@cad.zju.edu.cn

More information

Job Scheduling in a Distributed System Using Backfilling with Inaccurate Runtime Computations

Job Scheduling in a Distributed System Using Backfilling with Inaccurate Runtime Computations 2010 International Conference on Complex, Intelligent and Software Intensive Systems Job Scheduling in a Distributed System Using Backfilling with Inaccurate Runtime Computations Sofia K. Dimitriadou Department

More information

JOB/scheduling system is one of the core and challenging. An Greedy-Based Job Scheduling Algorithm in Cloud Computing

JOB/scheduling system is one of the core and challenging. An Greedy-Based Job Scheduling Algorithm in Cloud Computing JOURNAL OF SOFTWARE, VOL. 9, NO. 4, APRIL 2014 921 An Greedy-Based Job Scheduling Algorithm in Cloud Computing Ji Li a,b, Longhua Feng a,b, Shenglong Fang c a College of Computer Science, Chongqing University,

More information

Algorithmic Mechanism Design for Load Balancing in Distributed Systems

Algorithmic Mechanism Design for Load Balancing in Distributed Systems In Proc. of the 4th IEEE International Conference on Cluster Computing (CLUSTER 2002), September 24 26, 2002, Chicago, Illinois, USA, IEEE Computer Society Press, pp. 445 450. Algorithmic Mechanism Design

More information

High Availability Design Patterns

High Availability Design Patterns High Availability Design Patterns Kanwardeep Singh Ahluwalia 81-A, Punjabi Bagh, Patiala 147001 India kanwardeep@gmail.com +91 98110 16337 Atul Jain 135, Rishabh Vihar Delhi 110092 India jain.atul@wipro.com

More information

A Markovian Sensibility Analysis for Parallel Processing Scheduling on GNU/Linux

A Markovian Sensibility Analysis for Parallel Processing Scheduling on GNU/Linux A Markovian Sensibility Analysis for Parallel Processing Scheduling on GNU/Linux Regiane Y. Kawasaki 1, Luiz Affonso Guedes 2, Diego L. Cardoso 1, Carlos R. L. Francês 1, Glaucio H. S. Carvalho 1, Solon

More information

A Load Balancing Algorithm based on the Variation Trend of Entropy in Homogeneous Cluster

A Load Balancing Algorithm based on the Variation Trend of Entropy in Homogeneous Cluster , pp.11-20 http://dx.doi.org/10.14257/ ijgdc.2014.7.2.02 A Load Balancing Algorithm based on the Variation Trend of Entropy in Homogeneous Cluster Kehe Wu 1, Long Chen 2, Shichao Ye 2 and Yi Li 2 1 Beijing

More information

An Enhanced Cost Optimization of Heterogeneous Workload Management in Cloud Computing

An Enhanced Cost Optimization of Heterogeneous Workload Management in Cloud Computing An Enhanced Cost Optimization of Heterogeneous Workload Management in Cloud Computing 1 Sudha.C Assistant Professor/Dept of CSE, Muthayammal College of Engineering,Rasipuram, Tamilnadu, India Abstract:

More information

Job Scheduling with Moab Cluster Suite

Job Scheduling with Moab Cluster Suite Job Scheduling with Moab Cluster Suite IBM High Performance Computing February 2010 Y. Joanna Wong, Ph.D. yjw@us.ibm.com 2/22/2010 Workload Manager Torque Source: Adaptive Computing 2 Some terminology..

More information

AN ADAPTIVE DISTRIBUTED LOAD BALANCING TECHNIQUE FOR CLOUD COMPUTING

AN ADAPTIVE DISTRIBUTED LOAD BALANCING TECHNIQUE FOR CLOUD COMPUTING AN ADAPTIVE DISTRIBUTED LOAD BALANCING TECHNIQUE FOR CLOUD COMPUTING Gurpreet Singh M.Phil Research Scholar, Computer Science Dept. Punjabi University, Patiala gurpreet.msa@gmail.com Abstract: Cloud Computing

More information

Study of Knowledge Based Admission Control for a Software-as-a-Service Provider in Cloud Computing Environments

Study of Knowledge Based Admission Control for a Software-as-a-Service Provider in Cloud Computing Environments Study of Knowledge Based Admission Control for a Software-as-a-Service Provider in Cloud Computing Environments R.S.Mohana a, *, Dr. P. Thangaraj b,1, S.Kalaiselvi c,1 Abstract - Software as a Service

More information

Performance Gathering and Implementing Portability on Cloud Storage Data

Performance Gathering and Implementing Portability on Cloud Storage Data International Journal of Information & Computation Technology. ISSN 0974-2239 Volume 4, Number 17 (2014), pp. 1815-1823 International Research Publications House http://www. irphouse.com Performance Gathering

More information

Load balancing model for Cloud Data Center ABSTRACT:

Load balancing model for Cloud Data Center ABSTRACT: Load balancing model for Cloud Data Center ABSTRACT: Cloud data center management is a key problem due to the numerous and heterogeneous strategies that can be applied, ranging from the VM placement to

More information

RESOURCE CO-ALLOCATION ALGORITHMS IN DISTRIBUTED JOB BATCH SCHEDULING Victor V. Toporkov, Alexander Bobchenkov, Dmitry Yemelyanov and Anna Toporkova

RESOURCE CO-ALLOCATION ALGORITHMS IN DISTRIBUTED JOB BATCH SCHEDULING Victor V. Toporkov, Alexander Bobchenkov, Dmitry Yemelyanov and Anna Toporkova RESOURCE CO-ALLOCATION ALGORITHMS IN DISTRIBUTED JOB BATCH SCHEDULING Victor V. Toporkov, Alexander Bobchenkov, Dmitry Yemelyanov and Anna Toporkova Summary In this work, we present slot selection algorithms

More information

@IJMTER-2015, All rights Reserved 355

@IJMTER-2015, All rights Reserved 355 e-issn: 2349-9745 p-issn: 2393-8161 Scientific Journal Impact Factor (SJIF): 1.711 International Journal of Modern Trends in Engineering and Research www.ijmter.com A Model for load balancing for the Public

More information

Methodology of performance evaluation of integrated service systems with timeout control scheme

Methodology of performance evaluation of integrated service systems with timeout control scheme Methodology of performance evaluation of integrated service systems with timeout control scheme Akira Kawaguchi and Hiroshi Yamada NTT Service Integration Laboratories, NTT Corporation 9-11, Midori-cho

More information

Parallel Computing using MATLAB Distributed Compute Server ZORRO HPC

Parallel Computing using MATLAB Distributed Compute Server ZORRO HPC Parallel Computing using MATLAB Distributed Compute Server ZORRO HPC Goals of the session Overview of parallel MATLAB Why parallel MATLAB? Multiprocessing in MATLAB Parallel MATLAB using the Parallel Computing

More information

ENERGY-EFFICIENT TASK SCHEDULING ALGORITHMS FOR CLOUD DATA CENTERS

ENERGY-EFFICIENT TASK SCHEDULING ALGORITHMS FOR CLOUD DATA CENTERS ENERGY-EFFICIENT TASK SCHEDULING ALGORITHMS FOR CLOUD DATA CENTERS T. Jenifer Nirubah 1, Rose Rani John 2 1 Post-Graduate Student, Department of Computer Science and Engineering, Karunya University, Tamil

More information

The Probabilistic Model of Cloud Computing

The Probabilistic Model of Cloud Computing A probabilistic multi-tenant model for virtual machine mapping in cloud systems Zhuoyao Wang, Majeed M. Hayat, Nasir Ghani, and Khaled B. Shaban Department of Electrical and Computer Engineering, University

More information

A Review on Load Balancing In Cloud Computing 1

A Review on Load Balancing In Cloud Computing 1 www.ijecs.in International Journal Of Engineering And Computer Science ISSN:2319-7242 Volume 4 Issue 6 June 2015, Page No. 12333-12339 A Review on Load Balancing In Cloud Computing 1 Peenaz Pathak, 2 Er.Kamna

More information

Dynamic Resource management with VM layer and Resource prediction algorithms in Cloud Architecture

Dynamic Resource management with VM layer and Resource prediction algorithms in Cloud Architecture Dynamic Resource management with VM layer and Resource prediction algorithms in Cloud Architecture 1 Shaik Fayaz, 2 Dr.V.N.Srinivasu, 3 Tata Venkateswarlu #1 M.Tech (CSE) from P.N.C & Vijai Institute of

More information

DYNAMIC LOAD BALANCING IN A DECENTRALISED DISTRIBUTED SYSTEM

DYNAMIC LOAD BALANCING IN A DECENTRALISED DISTRIBUTED SYSTEM DYNAMIC LOAD BALANCING IN A DECENTRALISED DISTRIBUTED SYSTEM 1 Introduction In parallel distributed computing system, due to the lightly loaded and overloaded nodes that cause load imbalance, could affect

More information

Load Balancing on a Grid Using Data Characteristics

Load Balancing on a Grid Using Data Characteristics Load Balancing on a Grid Using Data Characteristics Jonathan White and Dale R. Thompson Computer Science and Computer Engineering Department University of Arkansas Fayetteville, AR 72701, USA {jlw09, drt}@uark.edu

More information

A taxonomy of market-based resource management systems for utility-driven cluster computing

A taxonomy of market-based resource management systems for utility-driven cluster computing SOFTWARE PRACTICE AND EXPERIENCE Softw. Pract. Exper. 2006; 36:1381 1419 Published online 8 June 2006 in Wiley InterScience (www.interscience.wiley.com)..725 A taxonomy of market-based resource management

More information

Network Model. University of Tsukuba. of the system. Load balancing policies are often. used for balancing the workload of distributed systems.

Network Model. University of Tsukuba. of the system. Load balancing policies are often. used for balancing the workload of distributed systems. CDC-INV A Performance Comparison of Dynamic vs. Static Load Balancing Policies in a Mainframe { Personal Computer Network Model Hisao Kameda El-Zoghdy Said Fathy y Inhwan Ryu z Jie Li x yzx University

More information

The Impact of Migration on Parallel Job. The Pennsylvania State University. University Park PA 16802. fyyzhang, anandg@cse.psu.edu. P. O.

The Impact of Migration on Parallel Job. The Pennsylvania State University. University Park PA 16802. fyyzhang, anandg@cse.psu.edu. P. O. The Impact of Migration on Parallel Job Scheduling for Distributed Systems Y. Zhang 1,H.Franke 2, J. E. Moreira 2, and A. Sivasubramaniam 1 1 Department of Computer Science & Engineering The Pennsylvania

More information

Resource Cost Optimization for Dynamic Load Balancing on Web Server System

Resource Cost Optimization for Dynamic Load Balancing on Web Server System Article can be accessed online at http://www.publishingindia.com Resource Cost Optimization for Dynamic Load Balancing on Web Server System Harikesh Singh*, Shishir Kumar** Abstract The growth of technology

More information

Performance Modeling and Analysis of a Database Server with Write-Heavy Workload

Performance Modeling and Analysis of a Database Server with Write-Heavy Workload Performance Modeling and Analysis of a Database Server with Write-Heavy Workload Manfred Dellkrantz, Maria Kihl 2, and Anders Robertsson Department of Automatic Control, Lund University 2 Department of

More information

A QoS-driven Resource Allocation Algorithm with Load balancing for

A QoS-driven Resource Allocation Algorithm with Load balancing for A QoS-driven Resource Allocation Algorithm with Load balancing for Device Management 1 Lanlan Rui, 2 Yi Zhou, 3 Shaoyong Guo State Key Laboratory of Networking and Switching Technology, Beijing University

More information

A Service Revenue-oriented Task Scheduling Model of Cloud Computing

A Service Revenue-oriented Task Scheduling Model of Cloud Computing Journal of Information & Computational Science 10:10 (2013) 3153 3161 July 1, 2013 Available at http://www.joics.com A Service Revenue-oriented Task Scheduling Model of Cloud Computing Jianguang Deng a,b,,

More information

Parallel Ray Tracing using MPI: A Dynamic Load-balancing Approach

Parallel Ray Tracing using MPI: A Dynamic Load-balancing Approach Parallel Ray Tracing using MPI: A Dynamic Load-balancing Approach S. M. Ashraful Kadir 1 and Tazrian Khan 2 1 Scientific Computing, Royal Institute of Technology (KTH), Stockholm, Sweden smakadir@csc.kth.se,

More information

Enabling Multi-pipeline Data Transfer in HDFS for Big Data Applications

Enabling Multi-pipeline Data Transfer in HDFS for Big Data Applications Enabling Multi-pipeline Data Transfer in HDFS for Big Data Applications Liqiang (Eric) Wang, Hong Zhang University of Wyoming Hai Huang IBM T.J. Watson Research Center Background Hadoop: Apache Hadoop

More information

Proposal and Development of a Reconfigurable Parallel Job Scheduling Algorithm

Proposal and Development of a Reconfigurable Parallel Job Scheduling Algorithm Proposal and Development of a Reconfigurable Parallel Job Scheduling Algorithm Luís Fabrício Wanderley Góes, Carlos Augusto Paiva da Silva Martins Graduate Program in Electrical Engineering PUC Minas {lfwgoes,capsm}@pucminas.br

More information

Mitglied der Helmholtz-Gemeinschaft. System monitoring with LLview and the Parallel Tools Platform

Mitglied der Helmholtz-Gemeinschaft. System monitoring with LLview and the Parallel Tools Platform Mitglied der Helmholtz-Gemeinschaft System monitoring with LLview and the Parallel Tools Platform November 25, 2014 Carsten Karbach Content 1 LLview 2 Parallel Tools Platform (PTP) 3 Latest features 4

More information