A Neural Network Approach for Dynamic Load Balancing In Homogeneous Distributed Systems

Size: px
Start display at page:

Download "A Neural Network Approach for Dynamic Load Balancing In Homogeneous Distributed Systems"

Transcription

1 A Neural Network Approach for Dynamic Balancing In Homogeneous Distributed Systems Aly E. El-Abd Department of Electronics and computer Engineering Arab Academy for Science and Technology P.O. Box 1029, Alexandria, Egypt Mohamed I. El-Bendary Department of Electronics and computer Engineering Arab Academy for Science and Technology P.O. Box 1029, Alexandria, Egypt Abstract A novel neural-based solution to the problem of dynamic load balancing in homogeneous distributed systems is proposed. The winner-take-all (WTA) neural network model is used for implementing the selection and location policies of a typical dynamic load balancing algorithm. Unlike most of the previous literature that assumed independent tasks, which is not always true, tasks with interprocess communication requirements are considered. All delays due to any usage of the communications network resource are taken into account. A simulation study was carried out to verify the effectiveness of the proposed approach, results were compared against the no load balancing case. Although performance improvements are dependent on the system overall load, load intensity per node, and nature of tasks, the results suggest that it is always beneficial to use load balancing than not at all. 1. Introduction The interest in building distributed systems has experienced a dramatic increase. Such increase was primarily driven by the vast advances and constantly declining costs of hardware technology and the various advantages these systems are known to provide such as resource sharing, high performance, and availability. This interest has spurred a large research activity in this area. One of the critical issues to the performance of distributed systems is the issue of balancing the work load of computational tasks among the different nodes comprising the system. imbalance is observed by the existence of nodes that are highly loaded while others are underloaded or even idle, such situations are harmfull to the system performance specially in terms of the average response time of tasks and the resourcestilization. The problem of load balancing has been tackled along several key dimensions. Static load balancing involves assigning tasks to processors at compile time. The problem has been formulated as an optimal assignment problem of tasks to processors [11,14,16,18]. Alternatively, the assignments can be made dynamically [1,4,5,12,18,20]. Dynamic load balancing involves the reallocation of tasks to processors after their initial assignments, which is done by migrating tasks from the overloaded nodes to other lightly loaded nodes so that load is shared evenly among the system nodes to improve the overall system performance. The problem of optimally making such assignment is NP-complete. Dynamic load balancing algorithms are characterized by the following key components: (i) The transfer policy: which decides when a task should be considered for migration, such decision is dependent on the current load level of the node. (ii) The selection policy: decides which task should be a candidate for migration. (iii) The location policy: decides to which host should a candidate task at an overloaded node be migrated. (iv) The information policy: which is the mechanism by which the load state information is exchanged between nodes, this may be done on periodic or event-driven basis. Since the decisions made by these policies are based on load information which is outdated due to the communication delay incurred in sending the load state update messages, we study the proposed policies taking all delay considerations into account. Recently, the location policy has received considerable attention [7,13]. However, very little on the selection policy have been reported in the open literature. Most of the previous work has chosen to trivialize the selection by either selecting the last task in queue or by selecting a random task. Other selection schemes were based on the service times of the tasks in the queue [21]. This was based on the assumption that all tasks are independent with no interprocess communication requirements. Instead, we consider both categories of tasks to exist, those independent and communicating tasks. The main focus of this work is the selection and location policies of dynamic load balancing algorithms.

2 The model we employ in our approach to implementing the selection and location policies is the Winner-Take-All (WTA) neural network model. This model is a single layer model, the connection weights between neurons are fixed so it is a fixed-weight net. The operation of this type of neural networks is based on competition [2], where neurons aggregate their inputs and the winner is the one with the largest net input, the winner neuron typically reinforces its output signal while simultaneously inhibiting the outputs of all other neurons in the layer. The motivation for using the WTA neural network is its simplicity and ease of implementation. The application of neural networks to the problem of dynamic load balancing has been reported in the literature by [8] where they have used a comparator neural network for the automated learning of load indices to be used by dynamic load balancing policies, although their work gives an insight to the area of application their results were under the unrealistic assumptions of no overheads or delays which are affecting factors of the quality of load balancing decisions. They have also integrated the comparator neural network with a population-based learning technique to automate the learning of load balancing policies[9]. The rest of this paper is organized as follows, section 2 describes the system model and discusses the assumptions concerning the load balancing strategy. Section 3 introduces the proposed approach for the selection and location policies. Section 4 discusses the simulation results and performance evaluation. Section 5 concludes this work. 2. System Model The aim of this section is defining the model of the distributed system under study, the host model, the task model, and the assumptions made Distributed System Model Our notion of a homogeneous distributed system is that of a multicomputer system. Communication between the system hosts is done solely by message passing. By homogeneity we mean that the hosts have the same software and hardware architecture, accordingly, any software component is assumed to be free to reside and being serviced on any host. We assume that the processing power may vary among the different hosts Host Model Each host in the system is assumed to have a computing facility (CPU), memory, and a disk. In developing the logical view of each host, we adopt the formulation by [12]. We assume that each host maintains two queues, the first queue is the active processes queue which houses the tasks that have started execution and have not yet finished. The second queue is the waiting processes queue which houses the tasks that have not yet started execution. Unlike most of the previous work, where the local scheduling policy assumed was the first-come-first-served (FCFS), we assume the local scheduling policy is a round-robin. The reasoning for this is that multiprogramming preemptive systems are much more efficient than their non-preemptive counterparts with a FCFS scheduling policy. Also the round-robin policy has an attractive feature over the FCFS policy which is that the waiting time of a task increases linearly with the task length [8]. Each host maintains load tables and routing tables as numerical values, ready to consume whenever the load balancing is invoked. balancing invocation is upon application arrival. Each host has a dedicated computing facility for executing the load balancing functions Task Model Most of the previous work has assumed that the tasks are independent, but this is not the real case all the way. Actually, jobs are often comprised of tasks that have interprocess communication requirements. A case that corresponds to a wide range of scientific problems. We assume that a job is partitioned at compile-time into several tasks with precedence constraints and explicit message-driven code. Initially, tasks are submitted to the waiting processes queue. A task is created and dispatched to the active processes queue if the following two conditions are met. First, the active processes queue is not at maximum capacity. Second, no precedence constraints are violated. An example of a task graph generated by the compiler is given in fig.(1) where ellipses correspond to tasks and arrows indicate precedence relations. The PARADIGM compilation system[3] developed at the university of Illinois is an example of the assumed parallelizing compiler. We define the following quantities used by the proposed selection and location policies.

3 X4 X2 Figure 1. Task graph example. P ij penalty assigned by task i to task j due to its need for communication with task j. S ij Communication Cost between task i and task j. E h (x) Estimated remaining execution time of task x at host h. T hk (x) Estimated cost of transfer of task x from host h to host k. U h Estimated unfinished work at host h. C hk Communication Cost / packet between hosts h and k. The proposed approach requires the existence of a compile and run-time libraries support for the estimation of execution times, remaining times, penalties, communication costs, and the unfinished work. A methodology that estimates the communication costs at compile time is reported in [3] based on the assumption that the underlying operating system supports communication primitives such as send-message and receive-message. The costs calculated are functions of the number of nodes involved in the communication (node-to-node, broadcast, or multicast) and the message length. Each penalty is the aggregated cost of all communication primitives between task i and task j. Hence, the communication cost between task i at host h and task j at host k, S ij, is given by: S ij = C hk *P ij In case of h=k, S ij is equal to zero. The estimated execution time of each task is obtained at compile-time. The estimated remaining execution time and estimated task transfer cost, say x from host h to host k are obtained at run-time. The estimated unfinished work is calculated at the time of load balancing invocation Assumptions X5 X1 X3 X6 Since the migration of a task after it has started execution incurs an enormous overhead that might outweigh any performance gains of load sharing, we assume that only non-preemptive transfers are allowed. (i.e., only tasks that have not yet started execution are considered for remote execution). The information policy adopted is the event-driven policy. A load state update message is sent by a host upon changing its load condition. The load conditions assumed are lightly loaded, normal load, and heavy loaded. We use two load measures, namely, the instantaneous queue length and the estimated unfinished work. The instantaneous queue length is the number of tasks existing in both the active and waiting processes queues. Using the queue length alone as the load measure and balancing parameter is not sufficient for obtaining good quality load balancing decisions, specially when variance in computing power at different hosts and variance of service times of tasks increase. The estimated unfinished work which is defined as the length of time needed for a host to complete all tasks in both of its queues. We adopt the acceptance policy proposed by [1] and recommended by [6] which is the request-reply policy allowing for wrong allocation decisions to be undone by giving the receiver host the right to refuse accepting more load. This also prevents the underloaded nodes from becoming heavily loaded due to excessive task transfers from several highly loaded nodes. The transfer policy assumed is a threshold policy, a host decides to cooperate in load balancing whenever its load index either falls below a certain threshold (becomes underloaded) or gets higher than an upper bound (becomes overloaded). 3. Proposed Selection and Location Policies This section discusses the proposed selection and location policies based on the winner-take-all (WTA) neural network model. Each of the two policies is mapped to a competitive layer neural network Selection Policy The WTA neural network employed has an architecture as depicted in figure (2). Each neuron represents a task of the waiting processes queue or the active processes queue. Each neuron has an associated bias that is equal to the estimated remaining execution time of the task at the local host. An inhibitory connection weight is set from node X i to node X j which is equal to the worst case communication cost per packet from the local host to any other host times the penalty P ij, which is the worst case S ij.. After the selection competitive layer is formed using the local information, the neurons compete for the right to be ON (fire). The winner neuron will denote the task with the largest difference between the estimated execution time and the interprocess communication cost. The reasoning behind this is the heuristic that the task with the highest execution time is more likely to be worth the transfer

4 overhead, and that the task with minimum interprocess communication requirements would yield a least cost transfer. Hence, the winner identifies the candidate task to be migrated. E 1 (x1) E 1 (x3) X1 X3 S 31 Figure 2. Selection policy neural network Location Policy S 21 S 41 S 32 S 43 X2 X4 S 42 E 1 (x2) E 1 (x4) The WTA neural network used for implementing the location policy has an architecture as shown in figure (3). Each neuron represents a host in the system. Each neuron accepts as input the estimated execution time of the selected task at the host represented by the neuron. Each neuron has an associated bias which is equal to the estimated unfinished work at the designated host. Connection weights are only from the neuron designating the local host to the other neurons representing hosts participating in load balancing. Weights values are equal to the estimated cost of transfer of the candidate task from the local host to each other participating host. All biases, inputs and weights are inhibitory since they represent load, execution and transfer overheads respectively. The location policy neural network is constructed based on the output of the selection layer and the current information concerning the load states in terms of the queue lengths of other hosts. In fact, the number of neurons in the location network is the number of the underloaded hosts of the system at the time of load balancing invocation. A host is eligible for being a receiver if its active processes queue length is less than its maximum length, so that a transferred task may start remote execution as soon as it arrives at the receiver. Once the network is constructed, neurons are allowed to compete for the ON activation level. The winner neuron identifies the candidate host to which the combined cost of execution and communication is minimum in accordance with available information about costs and loads. The candidate host is sent a request message for transferring the candidate task. The candidate host accepts the task by sending an accept message if it can still accommodate more load. If the load state of the candidate host has changed to normal or overload, or is currently waiting for tasks from other hosts that will result in load state change, a reject message is sent back to the local host. Both accept and reject messages will have the new load state information of the candidate host attached to them. In case of an accept message received, the task will be transferred immediately. In case of rejection, the load tables are examined, if there are other hosts that are eligible for participation in load balancing the competition phase is entered again. It should be noted that the winner neuron may be the one representing the local host, in such case the job will be executed locally in spite of the fact that there are other lightly loaded nodes in terms of the queue length alone. Additionally, we have incorporated the worst case communication cost between tasks so that a selected task will either be transferred to a host along the worst communication link or will be placed on a host to which less communication cost exists. Both previously stated features contribute in ensuring that good quality allocation decisions are made most of the time and that task transfer will only take place if it is beneficial. U 1 H1 H3 H4 T 43 (X*) U 4 U 3 Figure 3. Location policy neural net. 4. Performance Evaluation A simulation study was carried out in order to assess the performance of the proposed dynamic load balancing approach. We have used the mean response time as a performance criterion for evaluation. The simulation study assumes service times to be exponentially distributed and arrival rates to be poisson distributed. Although this may not hold in a practical sense, it would be sufficient in giving an overview of the performance[6]. In order to validate the simulation process, results from previous literature have been reproduced. 4.1 Qualitative Analysis U 2 T 41 (X*) H2 T 42 (X*) Qualitative analysis is concerned with the load

5 balancing algorithm properties such as generality, stability and scalability. The proposed selection and location policies are general enough to work with various transfer and information policies. A stable load sharing algorithm must prevent task thrashing by making the appropriate dynamic assignments of tasks to hosts so that the offered load gets serviced within a reasonable finite length of time. Scalability of a load sharing algorithm implies independence of physical system characteristics (e.g. system size, physical topology), and being tolerant to changes of physical resource characteristics (e.g. communication bandwidth and processor speed). Stability is a precondition for scalability [6]. The algorithm employs features that contribute to its stability, namely, the use of the request-reply policy which allows avoiding poor allocation decisions due to outdated information. Also, task thrashing is avoided since tasks are allowed to migrate only once as they are transferred to nodes where they can start execution immediately. The algorithm makes use of state information tables about all the nodes in the system, hence the algorithm depends on the number of nodes in the system so it is not scalable in this sense. Modifications to the algorithm are possible in order to account for scalability but they are not examined here. 4.2 Quantitative Analysis Quantitative analysis is concerned with the performance evaluation of the algorithm at different load conditions. We have modeled each host as an M/M/1 queuing system. The active processes queue has a maximum length of six tasks while the waiting processes queue is assumed to be of infinite capacity. A host is considered lightly loaded if its active processes queue has a length less than four. A host is considered in a normal load condition if its active queue has a length greater than four and less than six. A host is considered overloaded if its active queue is full and more than two tasks are getting queued in its waiting processes queue. The communication costs per packet assumed are shown in table (1). The algorithm is tested under different loads imposed on individual nodes but with the same overall load. We test the algorithm using tasks with interprocess communication for both cases of even and uneven load at the same overall load Light. The input data used for the light load testing is shown in tables(2, 3), for both cases of even and uneven loads. The results compared to the no load balancing case are also shown in figure (4). The numeral 1 on the x-axis denote the uneven load case while the numeral 2 denotes the even load case. The algorithm improves the mean response time of tasks substantially in both cases of even and uneven loads. An improvement of 34.38% is achieved in case of even load while in the case of uneven load the performance gain is around 47.73%. The reasoning behind this is that an uneven load pattern permits more task migration activity to take place due to the existence of several underloaded nodes. Response Time Pattern No Balancing Balancing Figure 4. Response time under light load of % Moderate. In the case of moderate load, two tests are carried out since this load condition is of prime interest. The first test is biased towards the light overall load condition while the second is biased towards the heavy overall load condition. In the first test the overall load was 70%. Results are shown in figure (5). The performance gain attained is around 41.36% for the case of the even load pattern while for the uneven load a performance improvement of 32% is achieved. An interesting result is that biasing the system towards the light load while imposing an uneven load pattern affects the performance gain negatively. This result is interpreted by a close examination of tables(4,5). Table(5) shows the uneven load pattern, although the overall load is still at the beginning of the spectrum of moderate loads (70%), the load distribution at the node is characterized by a high variance in the load intensities of nodes. Also the majority of nodes are at heavy load state (90%). This implies that a contention is expected to occur at the node of 10% intensity since it is highly underloaded. But the algorithm employs a request-reply policy which forces the highly loaded nodes to execute their load tasks locally whenever it can not accommodate more load. This increases the waiting time of tasks spent at the overloaded nodes. In contrast let us examine table(4) which shows the even load pattern case. All nodes are at the moderate load state, and the variance in the load intensities is substantially low. Such pattern allows for less but yet more effective task migration activity to occur. The second test shown in tables (6,7) is at the high end of the moderate loads spectrum. Results shown in figure (6) indicate that a performance improvement of 22.43% is achieved in the uneven load case while in the case of even load the improvement percentage was 21.03%. Clearly, in

6 both cases the performance gain is less than the previous tests due to imposing a higher overall load Response Time Response Time 1 Pattern 2 No Balance Balance Figure 5. Response time under moderate load of 70% Pattern No Balancing Balancing Figure 6. Response time under moderate load of 85% Heavy. The heavy load test is of a special interest because it is always important to get an estimate of the system performance under exceptionally heavy load conditions. Table (8) shows the test at an overall load of 95%. Performance is improved at this condition also which proves the robustness of the proposed load balancing scheme. The performance gain attained is around 21.03%. Common to all the tests is that the system overall throughput is kept nearly the same, this implies that the improvement in the response time does not hinder the system throughput which turns the performance gains achieved by this scheme into low-cost gains. Response Time Pattern No Balancing Balancing Figure 7. Response time under heavy load of 95%. Table 1. Communication costs/packet C12 C13 C14 C15 C23 C24 C25 C34 C35 C Table 2. Light load (41.16%-70%) even load pattern 1 Exponential(5s) Poisson(0.1176) 58.8% 2 Exponential(7s) Poisson(0.1) 70% 3 Exponential(6s) Poisson(0.111) 66% 4 Exponential(5s) Poisson(0.0588) 41.16% 5 Exponential(7s) Poisson(0.125) 62.5% % Table 3. Light load (15%-95%) uneven load pattern 1 Exponential(5s) Poisson(0.03) 15% 2 Exponential(7s) Poisson(0.04) 30% 3 Exponential(6s) Poisson(0.132) 79.23% 4 Exponential(5s) Poisson(0.113) 79.23% 5 Exponential(7s) Poisson(0.19) 95% % Table 4. Moderate load (65%-90%)even load pattern 1 Exponential(5s) Poisson(0.14) 70% 2 Exponential(7s) Poisson(0.093) 65% 3 Exponential(6s) Poisson(0.133) 80% 4 Exponential(5s) Poisson(0.14) 70% 5 Exponential(7s) Poisson(0.093) 65% 70% Table 5. Moderate load (10%-90%)uneven load pattern 1 Exponential(5s) Poisson(0.02) 10% 2 Exponential(7s) Poisson(0.128) 90% 3 Exponential(6s) Poisson(0.15) 90% 4 Exponential(5s) Poisson(0.14) 70% 5 Exponential(7s) Poisson(0.128) 90% 70% Table 6. Moderate load (85%-87.5%)even load pattern 1 Exponential(5s) Poisson(0.175) 87.5% 2 Exponential(7s) Poisson(0.121) 85% 3 Exponential(6s) Poisson(0.133) 80% 4 Exponential(5s) Poisson(0.121) 85% 5 Exponential(7s) Poisson(0.175) 87.5% 85%

7 Table 7. Moderate load (65%-90%)uneven load pattern 1 Exponential(5s) Poisson(0.13) 65% 2 Exponential(7s) Poisson(0.129) 90% 3 Exponential(6s) Poisson(0.15) 90% 4 Exponential(5s) Poisson(0.18) 90% 5 Exponential(7s) Poisson(0.129) 90% 85% Table 8. Heavy load (89%-98%)even load pattern 1 Exponential(5s) Poisson(0.19) 95% 2 Exponential(7s) Poisson(0.14) 98% 3 Exponential(6s) Poisson(0.148) 89% 4 Exponential(5s) Poisson(0.19) 95% 5 Exponential(7s) Poisson(0.14) 98% 95% 5. Conclusion and Future Work A novel solution to the problem of dynamic load balancing in a homogeneous distributed system based on the Winner-Take-All (WTA) neural network model was presented. Two different neural networks were introduced for implementing the selection and location policies of a typical dynamic load balancing algorithm. Both cases of independent tasks and tasks with interprocess communication requirements were considered during load balancing decision making. Also for achieving good quality allocation decisions, all communication costs associated with any use of the communications network resource were accounted for. The mean response time performance measure was found to improve for all cases but with different percentages. Several considerations are now under study. Modifications to the algorithm in order to be scalable, extending the approach to the heterogeneous case, and applying other neural network models specially those with learning capabilities, remain for future work. We hope that our work stimulates more research in this area in the near future. 6. References [1] D. Eager, E. Lazowska, and J. Zoharjan, Adaptive Sharing in Homogeneous Distributed Systems, IEEE Trans. Software Engineering, Vol.12, 1986, PP [3] M. Gupta and P. Banerjee, Compile-Time Estimation of Communication Costs on Multicomputers, Proc. Int l Parallel Processing Symposium, Beverly Hills, CA, March [4] A. Hac and X. Jin, Dynamic Balancing in a Distributed system using a sender-initiated Algorithm, Proc. 13 th Conf. Local Computer Network, [5] C.H. Hsu and J.W.-S. Liu, Dynamic Balancing Algorithms in Homogeneous Distributed Systems, Proc. 6 th Int l Conf. Distributed Computing Systems, [6] O. Kremien and J. Kramer, Methodical Analysis of Adaptive Sharing Algorithms, IEEE Trans. Parallel and Distributed Systems, Vol. 3, No. 6, 1992, PP [7] P. Krueger and N.G. Shivaratri, Adaptive Location Policies for Global Scheduling, IEEE Trans. Software Engineering, Vol. 20, No. 6, 1994, PP [8] P. Mehra and B. Wah, Automated Learning of Work Measures for Balancing on a Distributed System, Proc. Int l Conf. Parallel Processing, CRC Press, August, [9] P. Mehra and B. Wah, Population-Based Learning of Balancing Policies for a Distributed Computer System, Proc. Computing in Aerospace 9 Conf., AIAA, October, [10] R. Mirchandaney, D. Towsley, and J.A. Stankovic, Analysis of the Effects of Delays on Sharing, IEEE Trans. Computer, Vol.38, No.11, 1989, PP [11] G.S. Rao, H.S. Stone, and T.C. Hu, Assignment of tasks in a distributed processing system with limited memory, IEEE Trans. Computer, Vol. c-28, April 1979, PP [12] J.C. Ryou and J.S.K. Wong, A Task Migration Algorithm for Balancing in a Distributed System, Proc. 22 nd Annual Hawaii Int l Conf. System Sciences, Vol. 2, Software Track [13] K.G. Shin and Y. Chang, A Coordinated Location Policy for Sharing in Hypercube-Connected Multicomputers, IEEE Trans. Computers, Vol. 44, No.5, 1995, PP [14] K.G. Shin and M-S Chen, On the Number of Acceptable Task Assignments in Distributed Computing Systems, IEEE Trans. Computers, Vol. 39, No. 1, 1990, PP [15] J.A. Stankovic, Simulations of Three Adaptive, Decentralized controlled, Job Scheduling Algorithms, Computer Networks vol. 8, PP , August [16] H.S. Stone, S.H. Bokhari, Control of distributed processes, IEEE Computer, Vol. 11, 1978, PP [2] L. Fausett, Fundamentals of Neural Networks Architectures, Algorithms, and Applications, Prentice-Hall, inc., 1994.

8 [17]. S. Subramaniam and D.L. Eager, Affinity Scheduling of Unbalanced workloads, Proc. IEEE Supercomputing Conf., [18] A. Tantawi and D. Towsley, Optimal Static Balancing in Distributed Computer Systems, Journal ACM, Vol.32, April 1985, PP [19] Y. Wang and R.J.T. Morris, sharing in distributed Systems, IEEE Trans. Computer, Vol. C-34, No.3, 1985, PP [20] M.J. Zaki, W. Li, and S. Parthasarathy, Customized Dynamic Balancing for a Network of Workstations, Computer Science Dept., The University of Rochester, Technical Report 602, December [21] S. Zhou, A Trace-driven Simulations Study of Dynamic Balancing, IEEE Trans. Software Engineering, Vol.14, No.9, 1988, PP

Comparison on Different Load Balancing Algorithms of Peer to Peer Networks

Comparison on Different Load Balancing Algorithms of Peer to Peer Networks Comparison on Different Load Balancing Algorithms of Peer to Peer Networks K.N.Sirisha *, S.Bhagya Rekha M.Tech,Software Engineering Noble college of Engineering & Technology for Women Web Technologies

More information

Keywords: Dynamic Load Balancing, Process Migration, Load Indices, Threshold Level, Response Time, Process Age.

Keywords: Dynamic Load Balancing, Process Migration, Load Indices, Threshold Level, Response Time, Process Age. Volume 3, Issue 10, October 2013 ISSN: 2277 128X International Journal of Advanced Research in Computer Science and Software Engineering Research Paper Available online at: www.ijarcsse.com Load Measurement

More information

An Empirical Study and Analysis of the Dynamic Load Balancing Techniques Used in Parallel Computing Systems

An Empirical Study and Analysis of the Dynamic Load Balancing Techniques Used in Parallel Computing Systems An Empirical Study and Analysis of the Dynamic Load Balancing Techniques Used in Parallel Computing Systems Ardhendu Mandal and Subhas Chandra Pal Department of Computer Science and Application, University

More information

A Comparative Performance Analysis of Load Balancing Algorithms in Distributed System using Qualitative Parameters

A Comparative Performance Analysis of Load Balancing Algorithms in Distributed System using Qualitative Parameters A Comparative Performance Analysis of Load Balancing Algorithms in Distributed System using Qualitative Parameters Abhijit A. Rajguru, S.S. Apte Abstract - A distributed system can be viewed as a collection

More information

Various Schemes of Load Balancing in Distributed Systems- A Review

Various Schemes of Load Balancing in Distributed Systems- A Review 741 Various Schemes of Load Balancing in Distributed Systems- A Review Monika Kushwaha Pranveer Singh Institute of Technology Kanpur, U.P. (208020) U.P.T.U., Lucknow Saurabh Gupta Pranveer Singh Institute

More information

A Study on the Application of Existing Load Balancing Algorithms for Large, Dynamic, Heterogeneous Distributed Systems

A Study on the Application of Existing Load Balancing Algorithms for Large, Dynamic, Heterogeneous Distributed Systems A Study on the Application of Existing Load Balancing Algorithms for Large, Dynamic, Heterogeneous Distributed Systems RUPAM MUKHOPADHYAY, DIBYAJYOTI GHOSH AND NANDINI MUKHERJEE Department of Computer

More information

How To Compare Load Sharing And Job Scheduling In A Network Of Workstations

How To Compare Load Sharing And Job Scheduling In A Network Of Workstations A COMPARISON OF LOAD SHARING AND JOB SCHEDULING IN A NETWORK OF WORKSTATIONS HELEN D. KARATZA Department of Informatics Aristotle University of Thessaloniki 546 Thessaloniki, GREECE Email: karatza@csd.auth.gr

More information

RESEARCH PAPER International Journal of Recent Trends in Engineering, Vol 1, No. 1, May 2009

RESEARCH PAPER International Journal of Recent Trends in Engineering, Vol 1, No. 1, May 2009 An Algorithm for Dynamic Load Balancing in Distributed Systems with Multiple Supporting Nodes by Exploiting the Interrupt Service Parveen Jain 1, Daya Gupta 2 1,2 Delhi College of Engineering, New Delhi,

More information

Load Balancing Algorithms for Peer to Peer and Client Server Distributed Environments

Load Balancing Algorithms for Peer to Peer and Client Server Distributed Environments Load Balancing Algorithms for Peer to Peer and Client Server Distributed Environments Sameena Naaz Afshar Alam Ranjit Biswas Department of Computer Science Jamia Hamdard, New Delhi, India ABSTRACT Advancements

More information

A novel load balancing algorithm for computational grid

A novel load balancing algorithm for computational grid International Journal of Computational Intelligence Techniques, ISSN: 0976 0466 & E-ISSN: 0976 0474 Volume 1, Issue 1, 2010, PP-20-26 A novel load balancing algorithm for computational grid Saravanakumar

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

Comparative Study of Load Balancing Algorithms

Comparative Study of Load Balancing Algorithms IOSR Journal of Engineering (IOSRJEN) e-issn: 2250-3021, p-issn: 2278-8719 Vol. 3, Issue 3 (Mar. 2013), V2 PP 45-50 Comparative Study of Load Balancing Algorithms Jyoti Vashistha 1, Anant Kumar Jayswal

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

Performance Analysis of Load Balancing Algorithms in Distributed System

Performance Analysis of Load Balancing Algorithms in Distributed System Advance in Electronic and Electric Engineering. ISSN 2231-1297, Volume 4, Number 1 (2014), pp. 59-66 Research India Publications http://www.ripublication.com/aeee.htm Performance Analysis of Load Balancing

More information

Abstract. 1. Introduction

Abstract. 1. Introduction A REVIEW-LOAD BALANCING OF WEB SERVER SYSTEM USING SERVICE QUEUE LENGTH Brajendra Kumar, M.Tech (Scholor) LNCT,Bhopal 1; Dr. Vineet Richhariya, HOD(CSE)LNCT Bhopal 2 Abstract In this paper, we describe

More information

International journal of Engineering Research-Online A Peer Reviewed International Journal Articles available online http://www.ijoer.

International journal of Engineering Research-Online A Peer Reviewed International Journal Articles available online http://www.ijoer. RESEARCH ARTICLE ISSN: 2321-7758 GLOBAL LOAD DISTRIBUTION USING SKIP GRAPH, BATON AND CHORD J.K.JEEVITHA, B.KARTHIKA* Information Technology,PSNA College of Engineering & Technology, Dindigul, India Article

More information

International Journal of Scientific & Engineering Research, Volume 6, Issue 4, April-2015 36 ISSN 2229-5518

International Journal of Scientific & Engineering Research, Volume 6, Issue 4, April-2015 36 ISSN 2229-5518 International Journal of Scientific & Engineering Research, Volume 6, Issue 4, April-2015 36 An Efficient Approach for Load Balancing in Cloud Environment Balasundaram Ananthakrishnan Abstract Cloud computing

More information

Task Scheduling in Hadoop

Task Scheduling in Hadoop Task Scheduling in Hadoop Sagar Mamdapure Munira Ginwala Neha Papat SAE,Kondhwa SAE,Kondhwa SAE,Kondhwa Abstract Hadoop is widely used for storing large datasets and processing them efficiently under distributed

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

Load Balancing in the Cloud Computing Using Virtual Machine Migration: A Review

Load Balancing in the Cloud Computing Using Virtual Machine Migration: A Review Load Balancing in the Cloud Computing Using Virtual Machine Migration: A Review 1 Rukman Palta, 2 Rubal Jeet 1,2 Indo Global College Of Engineering, Abhipur, Punjab Technical University, jalandhar,india

More information

Dynamic Load Balancing Strategy for Grid Computing

Dynamic Load Balancing Strategy for Grid Computing Dynamic Load Balancing Strategy for Grid Computing Belabbas Yagoubi and Yahya Slimani Abstract Workload and resource management are two essential functions provided at the service level of the grid software

More information

A STUDY OF TASK SCHEDULING IN MULTIPROCESSOR ENVIROMENT Ranjit Rajak 1, C.P.Katti 2, Nidhi Rajak 3

A STUDY OF TASK SCHEDULING IN MULTIPROCESSOR ENVIROMENT Ranjit Rajak 1, C.P.Katti 2, Nidhi Rajak 3 A STUDY OF TASK SCHEDULING IN MULTIPROCESSOR ENVIROMENT Ranjit Rajak 1, C.P.Katti, Nidhi Rajak 1 Department of Computer Science & Applications, Dr.H.S.Gour Central University, Sagar, India, ranjit.jnu@gmail.com

More information

Load balancing in a heterogeneous computer system by self-organizing Kohonen network

Load balancing in a heterogeneous computer system by self-organizing Kohonen network Bull. Nov. Comp. Center, Comp. Science, 25 (2006), 69 74 c 2006 NCC Publisher Load balancing in a heterogeneous computer system by self-organizing Kohonen network Mikhail S. Tarkov, Yakov S. Bezrukov Abstract.

More information

Performance Evaluation of Mobile Agent-based Dynamic Load Balancing Algorithm

Performance Evaluation of Mobile Agent-based Dynamic Load Balancing Algorithm Performance Evaluation of Mobile -based Dynamic Load Balancing Algorithm MAGDY SAEB, CHERINE FATHY Computer Engineering Department Arab Academy for Science, Technology & Maritime Transport Alexandria,

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

A Review of Load Balancing Algorithms for Cloud Computing

A Review of Load Balancing Algorithms for Cloud Computing www.ijecs.in International Journal Of Engineering And Computer Science ISSN:2319-7242 Volume - 3 Issue -9 September, 2014 Page No. 8297-8302 A Review of Load Balancing Algorithms for Cloud Computing Dr.G.N.K.Sureshbabu

More information

Keywords Load balancing, Dispatcher, Distributed Cluster Server, Static Load balancing, Dynamic Load balancing.

Keywords Load balancing, Dispatcher, Distributed Cluster Server, Static Load balancing, Dynamic Load balancing. Volume 5, Issue 7, July 2015 ISSN: 2277 128X International Journal of Advanced Research in Computer Science and Software Engineering Research Paper Available online at: www.ijarcsse.com A Hybrid Algorithm

More information

How To Develop A Dynamic Load Balancing Algorithm

How To Develop A Dynamic Load Balancing Algorithm IJCSNS International Journal of Computer Science and Network Security, VOL.10 No.6, June 2010 153 A Guide to Dynamic Load Balancing in Distributed Computer Systems Ali M. Alakeel College of Computing and

More information

Improved Hybrid Dynamic Load Balancing Algorithm for Distributed Environment

Improved Hybrid Dynamic Load Balancing Algorithm for Distributed Environment International Journal of Scientific and Research Publications, Volume 3, Issue 3, March 2013 1 Improved Hybrid Dynamic Load Balancing Algorithm for Distributed Environment UrjashreePatil*, RajashreeShedge**

More information

LOAD BALANCING TECHNIQUES

LOAD BALANCING TECHNIQUES LOAD BALANCING TECHNIQUES Two imporatnt characteristics of distributed systems are resource multiplicity and system transparency. In a distributed system we have a number of resources interconnected by

More information

A Fuzzy Load Balancing Service for Network Computing Based on Jini

A Fuzzy Load Balancing Service for Network Computing Based on Jini A Fuzzy Load Balancing Service for Network Computing Based on Jini Lap-Sun Cheung and Yu-Kwong Kwok Department of Electrical and Electronic Engineering The University of Hong Kong, Pokfulam Road, Hong

More information

Improving Distributed Workload Performance by Sharing Both CPU and Memory Resources

Improving Distributed Workload Performance by Sharing Both CPU and Memory Resources Improving Distributed Workload Performance by Sharing Both CPU and Memory Resources Xiaodong Zhang Yanxia Qu Li Xiao Department of Computer Science College of William and Mary Williamsburg, VA 387-8795

More information

Sla Aware Load Balancing Algorithm Using Join-Idle Queue for Virtual Machines in Cloud Computing

Sla Aware Load Balancing Algorithm Using Join-Idle Queue for Virtual Machines in Cloud Computing Sla Aware Load Balancing Using Join-Idle Queue for Virtual Machines in Cloud Computing Mehak Choudhary M.Tech Student [CSE], Dept. of CSE, SKIET, Kurukshetra University, Haryana, India ABSTRACT: Cloud

More information

International Journal of Advancements in Research & Technology, Volume 3, Issue 8, August-2014 68 ISSN 2278-7763

International Journal of Advancements in Research & Technology, Volume 3, Issue 8, August-2014 68 ISSN 2278-7763 International Journal of Advancements in Research & Technology, Volume 3, Issue 8, August-2014 68 A Survey of Load Balancing Algorithms using VM B.KalaiSelvi 1 and Dr.L.Mary Immaculate Sheela 2 1 Research

More information

Load Balancing in Structured Peer to Peer Systems

Load Balancing in Structured Peer to Peer Systems Load Balancing in Structured Peer to Peer Systems DR.K.P.KALIYAMURTHIE 1, D.PARAMESWARI 2 Professor and Head, Dept. of IT, Bharath University, Chennai-600 073 1 Asst. Prof. (SG), Dept. of Computer Applications,

More information

Load Balancing in Structured Peer to Peer Systems

Load Balancing in Structured Peer to Peer Systems Load Balancing in Structured Peer to Peer Systems Dr.K.P.Kaliyamurthie 1, D.Parameswari 2 1.Professor and Head, Dept. of IT, Bharath University, Chennai-600 073. 2.Asst. Prof.(SG), Dept. of Computer Applications,

More information

DECENTRALIZED LOAD BALANCING IN HETEROGENEOUS SYSTEMS USING DIFFUSION APPROACH

DECENTRALIZED LOAD BALANCING IN HETEROGENEOUS SYSTEMS USING DIFFUSION APPROACH DECENTRALIZED LOAD BALANCING IN HETEROGENEOUS SYSTEMS USING DIFFUSION APPROACH P.Neelakantan Department of Computer Science & Engineering, SVCET, Chittoor pneelakantan@rediffmail.com ABSTRACT The grid

More information

A Survey on Load Balancing and Scheduling in Cloud Computing

A Survey on Load Balancing and Scheduling in Cloud Computing IJIRST International Journal for Innovative Research in Science & Technology Volume 1 Issue 7 December 2014 ISSN (online): 2349-6010 A Survey on Load Balancing and Scheduling in Cloud Computing Niraj Patel

More information

Scheduling and Load Balancing in Parallel and Distributed Systems

Scheduling and Load Balancing in Parallel and Distributed Systems Brochure More information from http://www.researchandmarkets.com/reports/2182702/ Scheduling and Load Balancing in Parallel and Distributed Systems Description: Advances in hardware and software technologies

More information

Fuzzy Active Queue Management for Assured Forwarding Traffic in Differentiated Services Network

Fuzzy Active Queue Management for Assured Forwarding Traffic in Differentiated Services Network Fuzzy Active Management for Assured Forwarding Traffic in Differentiated Services Network E.S. Ng, K.K. Phang, T.C. Ling, L.Y. Por Department of Computer Systems & Technology Faculty of Computer Science

More information

Dynamic Multi-User Load Balancing in Distributed Systems

Dynamic Multi-User Load Balancing in Distributed Systems Dynamic Multi-User Load Balancing in Distributed Systems Satish Penmatsa and Anthony T. Chronopoulos The University of Texas at San Antonio Dept. of Computer Science One UTSA Circle, San Antonio, Texas

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 SURVEY ON LOAD BALANCING ALGORITHMS FOR CLOUD COMPUTING

A SURVEY ON LOAD BALANCING ALGORITHMS FOR CLOUD COMPUTING A SURVEY ON LOAD BALANCING ALGORITHMS FOR CLOUD COMPUTING Avtar Singh #1,Kamlesh Dutta #2, Himanshu Gupta #3 #1 Department of Computer Science and Engineering, Shoolini University, avtarz@gmail.com #2

More information

International Journal of Computer Science Trends and Technology (IJCST) Volume 3 Issue 3, May-June 2015

International Journal of Computer Science Trends and Technology (IJCST) Volume 3 Issue 3, May-June 2015 RESEARCH ARTICLE OPEN ACCESS Ensuring Reliability and High Availability in Cloud by Employing a Fault Tolerance Enabled Load Balancing Algorithm G.Gayathri [1], N.Prabakaran [2] Department of Computer

More information

Key Words: Dynamic Load Balancing, and Distributed System

Key Words: Dynamic Load Balancing, and Distributed System DYNAMIC ROTATING LOAD BALANCING ALGORITHM IN DISTRIBUTED SYSTEMS ROSE SULEIMAN AL DAHOUD ALI ISSA OTOUM Al-Zaytoonah University Al-Zaytoonah University Neelain University rosesuleiman@yahoo.com aldahoud@alzaytoonah.edu.jo

More information

Performance advantages of resource sharing in polymorphic optical networks

Performance advantages of resource sharing in polymorphic optical networks R. J. Durán, I. de Miguel, N. Merayo, P. Fernández, R. M. Lorenzo, E. J. Abril, I. Tafur Monroy, Performance advantages of resource sharing in polymorphic optical networks, Proc. of the 0th European Conference

More information

LOAD BALANCING ALGORITHMS IN CLOUD COMPUTING

LOAD BALANCING ALGORITHMS IN CLOUD COMPUTING International Journal of Advanced Computer and Mathematical Sciences ISSN 2230-9624. Vol4, Issue3, 2013, pp229-233 http://bipublication.com LOAD BALANCING ALGORITHMS IN CLOUD COMPUTING Doddini Probhuling

More information

Study of Various Load Balancing Techniques in Cloud Environment- A Review

Study of Various Load Balancing Techniques in Cloud Environment- A Review International Journal of Computer Sciences and Engineering Open Access Review Paper Volume-4, Issue-04 E-ISSN: 2347-2693 Study of Various Load Balancing Techniques in Cloud Environment- A Review Rajdeep

More information

Load Balancing in Distributed System. Prof. Ananthanarayana V.S. Dept. Of Information Technology N.I.T.K., Surathkal

Load Balancing in Distributed System. Prof. Ananthanarayana V.S. Dept. Of Information Technology N.I.T.K., Surathkal Load Balancing in Distributed System Prof. Ananthanarayana V.S. Dept. Of Information Technology N.I.T.K., Surathkal Objectives of This Module Show the differences between the terms CPU scheduling, Job

More information

Load Balancing of Web Server System Using Service Queue Length

Load Balancing of Web Server System Using Service Queue Length Load Balancing of Web Server System Using Service Queue Length Brajendra Kumar 1, Dr. Vineet Richhariya 2 1 M.tech Scholar (CSE) LNCT, Bhopal 2 HOD (CSE), LNCT, Bhopal Abstract- In this paper, we describe

More information

Dept. of Computer Engg., Bhartia Vidyapeeth, Pune, India

Dept. of Computer Engg., Bhartia Vidyapeeth, Pune, India The Server Reassignment Problem for Load Balancing In Structured Peer to Peer Systems 1 Sumit A. Hirve, 2 Dr. S.H. Patil 1,2 Dept. of Computer Engg., Bhartia Vidyapeeth, Pune, India Abstract Application-layer

More information

Load Balancing In Concurrent Parallel Applications

Load Balancing In Concurrent Parallel Applications Load Balancing In Concurrent Parallel Applications Jeff Figler Rochester Institute of Technology Computer Engineering Department Rochester, New York 14623 May 1999 Abstract A parallel concurrent application

More information

There are a number of factors that increase the risk of performance problems in complex computer and software systems, such as e-commerce systems.

There are a number of factors that increase the risk of performance problems in complex computer and software systems, such as e-commerce systems. ASSURING PERFORMANCE IN E-COMMERCE SYSTEMS Dr. John Murphy Abstract Performance Assurance is a methodology that, when applied during the design and development cycle, will greatly increase the chances

More information

A Comparison of Dynamic Load Balancing Algorithms

A Comparison of Dynamic Load Balancing Algorithms A Comparison of Dynamic Load Balancing Algorithms Toufik Taibi 1, Abdelouahab Abid 2 and Engku Fariez Engku Azahan 2 1 College of Information Technology, United Arab Emirates University, P.O. Box 17555,

More information

Comparison of Request Admission Based Performance Isolation Approaches in Multi-tenant SaaS Applications

Comparison of Request Admission Based Performance Isolation Approaches in Multi-tenant SaaS Applications Comparison of Request Admission Based Performance Isolation Approaches in Multi-tenant SaaS Applications Rouven Kreb 1 and Manuel Loesch 2 1 SAP AG, Walldorf, Germany 2 FZI Research Center for Information

More information

Energy Constrained Resource Scheduling for Cloud Environment

Energy Constrained Resource Scheduling for Cloud Environment Energy Constrained Resource Scheduling for Cloud Environment 1 R.Selvi, 2 S.Russia, 3 V.K.Anitha 1 2 nd Year M.E.(Software Engineering), 2 Assistant Professor Department of IT KSR Institute for Engineering

More information

An Approach to Load Balancing In Cloud Computing

An Approach to Load Balancing In Cloud Computing An Approach to Load Balancing In Cloud Computing Radha Ramani Malladi Visiting Faculty, Martins Academy, Bangalore, India ABSTRACT: Cloud computing is a structured model that defines computing services,

More information

Scheduling and Load Balancing in Parallel and Distributed Systems

Scheduling and Load Balancing in Parallel and Distributed Systems Scheduling and Load Balancing in Parallel and Distributed Systems Behrooz A. Shirazi Ali R. Hurson Krishna M.Kayi IEEE Computer Society Press Los Alamitos, California Washington Brussels Tokyo - a?-/ Contents

More information

How To Balance In Cloud Computing

How To Balance In Cloud Computing A Review on Load Balancing Algorithms in Cloud Hareesh M J Dept. of CSE, RSET, Kochi hareeshmjoseph@ gmail.com John P Martin Dept. of CSE, RSET, Kochi johnpm12@gmail.com Yedhu Sastri Dept. of IT, RSET,

More information

Reallocation and Allocation of Virtual Machines in Cloud Computing Manan D. Shah a, *, Harshad B. Prajapati b

Reallocation and Allocation of Virtual Machines in Cloud Computing Manan D. Shah a, *, Harshad B. Prajapati b Proceedings of International Conference on Emerging Research in Computing, Information, Communication and Applications (ERCICA-14) Reallocation and Allocation of Virtual Machines in Cloud Computing Manan

More information

Load Distribution in Large Scale Network Monitoring Infrastructures

Load Distribution in Large Scale Network Monitoring Infrastructures Load Distribution in Large Scale Network Monitoring Infrastructures Josep Sanjuàs-Cuxart, Pere Barlet-Ros, Gianluca Iannaccone, and Josep Solé-Pareta Universitat Politècnica de Catalunya (UPC) {jsanjuas,pbarlet,pareta}@ac.upc.edu

More information

Dynamic Load Balancing in a Network of Workstations

Dynamic Load Balancing in a Network of Workstations Dynamic Load Balancing in a Network of Workstations 95.515F Research Report By: Shahzad Malik (219762) November 29, 2000 Table of Contents 1 Introduction 3 2 Load Balancing 4 2.1 Static Load Balancing

More information

A Clustered Approach for Load Balancing in Distributed Systems

A Clustered Approach for Load Balancing in Distributed Systems SSRG International Journal of Mobile Computing & Application (SSRG-IJMCA) volume 2 Issue 1 Jan to Feb 2015 A Clustered Approach for Load Balancing in Distributed Systems Shweta Rajani 1, Niharika Garg

More information

Using Fuzzy Logic Control to Provide Intelligent Traffic Management Service for High-Speed Networks ABSTRACT:

Using Fuzzy Logic Control to Provide Intelligent Traffic Management Service for High-Speed Networks ABSTRACT: Using Fuzzy Logic Control to Provide Intelligent Traffic Management Service for High-Speed Networks ABSTRACT: In view of the fast-growing Internet traffic, this paper propose a distributed traffic management

More information

A Survey on Load Balancing Algorithms in Cloud Environment

A Survey on Load Balancing Algorithms in Cloud Environment A Survey on Load s in Cloud Environment M.Aruna Assistant Professor (Sr.G)/CSE Erode Sengunthar Engineering College, Thudupathi, Erode, India D.Bhanu, Ph.D Associate Professor Sri Krishna College of Engineering

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

CHAPTER 5 WLDMA: A NEW LOAD BALANCING STRATEGY FOR WAN ENVIRONMENT

CHAPTER 5 WLDMA: A NEW LOAD BALANCING STRATEGY FOR WAN ENVIRONMENT 81 CHAPTER 5 WLDMA: A NEW LOAD BALANCING STRATEGY FOR WAN ENVIRONMENT 5.1 INTRODUCTION Distributed Web servers on the Internet require high scalability and availability to provide efficient services to

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

A Dynamic Approach for Load Balancing using Clusters

A Dynamic Approach for Load Balancing using Clusters A Dynamic Approach for Load Balancing using Clusters ShwetaRajani 1, RenuBagoria 2 Computer Science 1,2,Global Technical Campus, Jaipur 1,JaganNath University, Jaipur 2 Email: shwetarajani28@yahoo.in 1

More information

Load Balancing in Computer Networks

Load Balancing in Computer Networks Load Balancing in Computer Networks Ming-Chang Huang, S. Hossein Hosseini 1 and K. Vairavan Department of Electrical Engineering and Computer Science University of Wisconsin Milwaukee PO Box 784 Milwaukee,

More information

Adaptive Load Balancing Method Enabling Auto-Specifying Threshold of Node Load Status for Apache Flume

Adaptive Load Balancing Method Enabling Auto-Specifying Threshold of Node Load Status for Apache Flume , pp. 201-210 http://dx.doi.org/10.14257/ijseia.2015.9.2.17 Adaptive Load Balancing Method Enabling Auto-Specifying Threshold of Node Load Status for Apache Flume UnGyu Han and Jinho Ahn Dept. of Comp.

More information

In Proceedings of the First IEEE Workshop on Real-Time Applications, New York, NY, May 1993.

In Proceedings of the First IEEE Workshop on Real-Time Applications, New York, NY, May 1993. In Proceedings of the First IEEE Workshop on Real-Time Applications, New York, NY, May 1993. Probabilistic Job Scheduling for Distributed Real-time Applications Azer Bestavros Dimitrios Spartiotis Computer

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

Load Balancing Between Heterogenous Computing Clusters

Load Balancing Between Heterogenous Computing Clusters Load Balancing Between Heterogenous Computing Clusters Siu-Cheung Chau Dept. of Physics and Computing, Wilfrid Laurier University, Waterloo, Ontario, Canada, N2L 3C5 e-mail: schau@wlu.ca Ada Wai-Chee Fu

More information

Roulette Wheel Selection Model based on Virtual Machine Weight for Load Balancing in Cloud Computing

Roulette Wheel Selection Model based on Virtual Machine Weight for Load Balancing in Cloud Computing IOSR Journal of Computer Engineering (IOSR-JCE) e-issn: 2278-0661,p-ISSN: 2278-8727, Volume 16, Issue 5, Ver. VII (Sep Oct. 2014), PP 65-70 Roulette Wheel Selection Model based on Virtual Machine Weight

More information

Path Selection Methods for Localized Quality of Service Routing

Path Selection Methods for Localized Quality of Service Routing Path Selection Methods for Localized Quality of Service Routing Xin Yuan and Arif Saifee Department of Computer Science, Florida State University, Tallahassee, FL Abstract Localized Quality of Service

More information

Group Based Load Balancing Algorithm in Cloud Computing Virtualization

Group Based Load Balancing Algorithm in Cloud Computing Virtualization Group Based Load Balancing Algorithm in Cloud Computing Virtualization Rishi Bhardwaj, 2 Sangeeta Mittal, Student, 2 Assistant Professor, Department of Computer Science, Jaypee Institute of Information

More information

Preserving Message Integrity in Dynamic Process Migration

Preserving Message Integrity in Dynamic Process Migration Preserving Message Integrity in Dynamic Process Migration E. Heymann, F. Tinetti, E. Luque Universidad Autónoma de Barcelona Departamento de Informática 8193 - Bellaterra, Barcelona, Spain e-mail: e.heymann@cc.uab.es

More information

Use of Agent-Based Service Discovery for Resource Management in Metacomputing Environment

Use of Agent-Based Service Discovery for Resource Management in Metacomputing Environment In Proceedings of 7 th International Euro-Par Conference, Manchester, UK, Lecture Notes in Computer Science 2150, Springer Verlag, August 2001, pp. 882-886. Use of Agent-Based Service Discovery for Resource

More information

Proposal of Dynamic Load Balancing Algorithm in Grid System

Proposal of Dynamic Load Balancing Algorithm in Grid System www.ijcsi.org 186 Proposal of Dynamic Load Balancing Algorithm in Grid System Sherihan Abu Elenin Faculty of Computers and Information Mansoura University, Egypt Abstract This paper proposed dynamic load

More information

Design of an Optimized Virtual Server for Efficient Management of Cloud Load in Multiple Cloud Environments

Design of an Optimized Virtual Server for Efficient Management of Cloud Load in Multiple Cloud Environments Design of an Optimized Virtual Server for Efficient Management of Cloud Load in Multiple Cloud Environments Ajay A. Jaiswal 1, Dr. S. K. Shriwastava 2 1 Associate Professor, Department of Computer Technology

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

A Review of Customized Dynamic Load Balancing for a Network of Workstations

A Review of Customized Dynamic Load Balancing for a Network of Workstations A Review of Customized Dynamic Load Balancing for a Network of Workstations Taken from work done by: Mohammed Javeed Zaki, Wei Li, Srinivasan Parthasarathy Computer Science Department, University of Rochester

More information

Load Balancing in cloud computing

Load Balancing in cloud computing Load Balancing in cloud computing 1 Foram F Kherani, 2 Prof.Jignesh Vania Department of computer engineering, Lok Jagruti Kendra Institute of Technology, India 1 kheraniforam@gmail.com, 2 jigumy@gmail.com

More information

A Survey Of Various Load Balancing Algorithms In Cloud Computing

A Survey Of Various Load Balancing Algorithms In Cloud Computing A Survey Of Various Load Balancing Algorithms In Cloud Computing Dharmesh Kashyap, Jaydeep Viradiya Abstract: Cloud computing is emerging as a new paradigm for manipulating, configuring, and accessing

More information

Using median filtering in active queue management for telecommunication networks

Using median filtering in active queue management for telecommunication networks Using median filtering in active queue management for telecommunication networks Sorin ZOICAN *, Ph.D. Cuvinte cheie. Managementul cozilor de aşteptare, filtru median, probabilitate de rejectare, întârziere.

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

Optical Network Traffic Control Algorithm under Variable Loop Delay: A Simulation Approach

Optical Network Traffic Control Algorithm under Variable Loop Delay: A Simulation Approach Int. J. Communications, Network and System Sciences, 2009, 7, 652-656 doi:10.4236/icns.2009.27074 Published Online October 2009 (http://www.scirp.org/ournal/icns/). Optical Network Traffic Control Algorithm

More information

International Journal of Advance Research in Computer Science and Management Studies

International Journal of Advance Research in Computer Science and Management Studies Volume 3, Issue 6, June 2015 ISSN: 2321 7782 (Online) International Journal of Advance Research in Computer Science and Management Studies Research Article / Survey Paper / Case Study Available online

More information

Performance Analysis of VM Scheduling Algorithm of CloudSim in Cloud Computing

Performance Analysis of VM Scheduling Algorithm of CloudSim in Cloud Computing IJECT Vo l. 6, Is s u e 1, Sp l-1 Ja n - Ma r c h 2015 ISSN : 2230-7109 (Online) ISSN : 2230-9543 (Print) Performance Analysis Scheduling Algorithm CloudSim in Cloud Computing 1 Md. Ashifuddin Mondal,

More information

Flexible Deterministic Packet Marking: An IP Traceback Scheme Against DDOS Attacks

Flexible Deterministic Packet Marking: An IP Traceback Scheme Against DDOS Attacks Flexible Deterministic Packet Marking: An IP Traceback Scheme Against DDOS Attacks Prashil S. Waghmare PG student, Sinhgad College of Engineering, Vadgaon, Pune University, Maharashtra, India. prashil.waghmare14@gmail.com

More information

A Hybrid Load Balancing Policy underlying Cloud Computing Environment

A Hybrid Load Balancing Policy underlying Cloud Computing Environment A Hybrid Load Balancing Policy underlying Cloud Computing Environment S.C. WANG, S.C. TSENG, S.S. WANG*, K.Q. YAN* Chaoyang University of Technology 168, Jifeng E. Rd., Wufeng District, Taichung 41349

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

Efficient DNS based Load Balancing for Bursty Web Application Traffic

Efficient DNS based Load Balancing for Bursty Web Application Traffic ISSN Volume 1, No.1, September October 2012 International Journal of Science the and Internet. Applied However, Information this trend leads Technology to sudden burst of Available Online at http://warse.org/pdfs/ijmcis01112012.pdf

More information

Public Cloud Partition Balancing and the Game Theory

Public Cloud Partition Balancing and the Game Theory Statistics Analysis for Cloud Partitioning using Load Balancing Model in Public Cloud V. DIVYASRI 1, M.THANIGAVEL 2, T. SUJILATHA 3 1, 2 M. Tech (CSE) GKCE, SULLURPETA, INDIA v.sridivya91@gmail.com thaniga10.m@gmail.com

More information

APPENDIX 1 USER LEVEL IMPLEMENTATION OF PPATPAN IN LINUX SYSTEM

APPENDIX 1 USER LEVEL IMPLEMENTATION OF PPATPAN IN LINUX SYSTEM 152 APPENDIX 1 USER LEVEL IMPLEMENTATION OF PPATPAN IN LINUX SYSTEM A1.1 INTRODUCTION PPATPAN is implemented in a test bed with five Linux system arranged in a multihop topology. The system is implemented

More information

Programma della seconda parte del corso

Programma della seconda parte del corso Programma della seconda parte del corso Introduction Reliability Performance Risk Software Performance Engineering Layered Queueing Models Stochastic Petri Nets New trends in software modeling: Metamodeling,

More information

CDBMS Physical Layer issue: Load Balancing

CDBMS Physical Layer issue: Load Balancing CDBMS Physical Layer issue: Load Balancing Shweta Mongia CSE, School of Engineering G D Goenka University, Sohna Shweta.mongia@gdgoenka.ac.in Shipra Kataria CSE, School of Engineering G D Goenka 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