Dynamic Load Balancing Policy Research Articles

Dynamic virtual machine allocation policy in cloud computing complying with service level agreement using CloudSim

Cloud computing provides services over the internet and provides application resources and data to the users based on their demand. Base of the Cloud Computing is consumer provider model. Cloud provider provides resources which consumer can access using cloud computing model in order to build their application based on their demand. Cloud data center is a bulk of resources on shared pool architecture for cloud user to access. Virtualization is the heart of the Cloud computing model, it provides virtual machine as per application specific configuration and those applications are free to choose their own configuration. On one hand, there is huge number of resources and on other hand it has to serve huge number of requests effectively. Therefore, resource allocation policy and scheduling policy play very important role in allocation and managing resources in this cloud computing model. This paper proposes the load balancing policy using Hungarian algorithm. Hungarian Algorithm provides dynamic load balancing policy with a monitor component. Monitor component helps to increase cloud resource utilization by managing the Hungarian algorithm by monitoring its state and altering its state based on artificial intelligent. CloudSim used in this proposal is an extensible toolkit and it simulates cloud computing environment.

Delay‐ and energy‐aware load balancing in ultra‐dense heterogeneous 5G networks

AbstractIn the context of ultra‐dense heterogeneous networks in 5G, we focus on the load balancing problem of elastic traffic in a single macrocell that is supported by a number of small cells. Each small cell is modeled by a single‐class processor sharing queue, whereas the macrocell is modelled by a multiclass processor sharing queue. We assume that the macrocell is always consuming energy, while a small cell can be switched off when it is idle with the setup delay penalty. As the main contribution, aimed at minimizing the weighted sum of the mean delay and the mean power consumption in the system, we develop static and dynamic load balancing policies that take into account the setup delay. All the obtained results are analytically founded. The performance of the static policy is evaluated using the achieved analytical expressions, which include the static optimal probabilities achieved before, whereas for the dynamic policy we need to perform simulations using previously obtained numerical results from the first policy iteration procedure. As concluded, numerical results indicate that the dynamic policy outperforms the static policy under realistic settings.

Load splitting in clusters of video servers

Nowadays, video on demand is one of the services more highly appreciated and demanded by customers. As the number of users increases, the capacity of the system that provides these services must also be increased to guarantee the required quality of service. An approach to that end is to have available several video servers at various distribution points in order to satisfy the different incoming demands (video server cluster). When a movie demand arrives to such a cluster, a load balancing device must assign the request to a specific server according to a procedure that must be fast, easy to implement and scalable. In this article we consider the problem of appropriately splitting this load to improve on the system performance. After an analysis of the video packet generation, we point out the similarity between this problem and that of optimally routing packets in data networks. With this similarity in mind, a new mechanism to select the appropriate video server is proposed. The purpose of this mechanism is to minimize the average packet transfer time (waiting time plus transmission time) at the video server cluster. In this way, we are able to obtain a dynamic load balancing policy that performs satisfactorily and that is very easy to implement in practice. The results of several experiments run with real data are shown and commented to substantiate our claims. A description of a practical implementation of the system is also included.

Towards a hybrid load balancing policy in grid computing system

Grid computing has become conventional in distributed systems due to technological advancements and network popularity. Grid computing facilitates distributed applications by integrating available idle network computing resources into formidable computing power. As a result, by using efficient integration and sharing of resources, this enables abundant computing resources to solve complicated problems that a single machine cannot manage. However, grid computing mines resources from accessible idle nodes and node accessibility varies with time. A node that is currently idle, may become occupied within a second of time and then be unavailable to provide resources. Accordingly, node selection must provide effective and sufficient resources over a long period to allow load assignment. This study proposes a hybrid load balancing policy to integrate static and dynamic load balancing technologies. Essentially, a static load balancing policy is applied to select effective and suitable node sets. This will lower the unbalanced load probability caused by assigning tasks to ineffective nodes. When a node reveals the possible inability to continue providing resources, the dynamic load balancing policy will determine whether the node in question is ineffective to provide load assignment. The system will then obtain a new replacement node within a short time, to maintain system execution performance.

Dynamic Load Balancing in Distributed Systems in the Presence of Delays: A Regeneration-Theory Approach

A regeneration-theory approach is undertaken to analytically characterize the average overall completion time in a distributed system. The approach considers the heterogeneity in the processing rates of the nodes as well as the randomness in the delays imposed by the communication medium. The optimal one-shot load balancing policy is developed and subsequently extended to develop an autonomous and distributed load-balancing policy that can dynamically reallocate incoming external loads at each node. This adaptive and dynamic load balancing policy is implemented and evaluated in a two-node distributed system. The performance of the proposed dynamic load-balancing policy is compared to that of static policies as well as existing dynamic load-balancing policies by considering the average completion time per task and the system processing rate in the presence of random arrivals of the external loads.

Decentralized load balancing for highly irregular search problems

In this paper, we present a distributed computing framework for problems characterized by a highly irregular search tree, whereby no reliable workload prediction is available. The framework is based on a peer-to-peer computing environment and dynamic load balancing. The system allows for dynamic resource aggregation, does not depend on any specific meta-computing middleware and is suitable for large-scale, multi-domain, heterogeneous environments, such as computational Grids. Dynamic load balancing policies based on global statistics are known to provide optimal load balancing performance, while randomized techniques provide high scalability. The proposed method combines both advantages and adopts distributed job-pools and a randomized polling technique. The framework has been successfully adopted in a parallel search algorithm for subgraph mining and evaluated on a molecular compounds dataset. The parallel application has shown good scalability and close-to linear speedup in a distributed network of workstations.

Threshold-Based Load Balancing for Multiple Home Agents in Mobile IP Networks

Mobile IP is an extension of the IP protocol, designed to provide seamless connectivity to mobile nodes roaming over the Internet. Under certain conditions of traffic, such as in Mobile IP networks supporting multimedia applications, overhead can cause significant delays at the mobility agents, i.e. foreign and home agents. Some multiple home agents protocol extensions have been proposed to avoid potential bottlenecks in single home agents configurations. Due to the highly unpredictable nature of the IP data traffic, efficient load balancing applied in such schemes is crucial to the overall performance improvement. In this paper, we propose a novel threshold-based dynamic load balancing policy that uses a realistic data-sharing model operational at each home agent. A discrete event computer simulator is developed to study the system performance with the proposed load balancing policy. We also introduce two new policy tuning parameters, the number of information queue slices and the retry latency, by which we control the load balancing benefit. Due to the relatively large number of simulation parameters, results are presented for typical configuration scenarios. In general, in all the results obtained, our threshold-based load balancing policy provides a significant performance improvement. The proposed architecture efficiently uses parallel processing for handling the overhead generated by the load balancing, which results in minimal performance degradation.

A Combined Decomposition and State Aggregation Method for Analyzing Dynamic Load-Balancing Policies

Exact performance analysis of dynamic load-balancing policies for distributed systems is known to be very difficult due to state dependent load-balancing decisions and the large size of state spaces. Therefore, approximate methods are usually used. There have been two useful methods for analyzing dynamic load-balancing policies, namely, the state-aggregation method and the decomposition method. This paper proposes a method which uses a combination of these two methods and introduces a new concept called a correlation window for analyzing dynamic load-balancing policies.

Automated learning of load-balancing strategies in multiprogrammed distributed systems

Dynamic load-balancing strategies for distributed systems seek to improve average completion time of independent tasks by migrating each incoming task to the site where it is expected to finish the fastest: usually the site having the smallest load index. SMALL is an offline learning system for developing configuration-specific load-balancing strategies; it learns new load indices as well as tunes the parameters of given migration policies. Using a dynamic workload generator, a number of typical systemwide load patterns are first recorded; the completion times of several benchmark jobs are then measured at each site, under each of the recorded load patterns. These measurements are used to train comparator neural networks simultaneously, one per site. The comparators collectively model a set of perfect load indices in that they seek to rank, at arrival time, the possible destinations for an incoming task by their (not yet known) respective completion times. The numerous parameters of the decentralized dynamic load-balancing policy are then tuned using a genetic algorithm. We present experimental results for a mix of scientific and interactive workloads on Sun workstations connected by Ethernet. The policies tuned by SMALL are shown to exploit idle resources intelligently and effectively.

Modelling And Analyses Of Dynamic Load-Balancing Policies By State Aggregation

Exact performance analysis of dynamic load-balancing policies for distributed systems is known to be very difficult owing to the facts that the state space is multidimensional and that load-balancing decisions are state dependent. In this paper, a state-aggregation method is proposed to model and analyze dynamic load-balancing policies. Those states with the same number of jobs are aggregated into a single state. The number of jobs in the system is modelled by a birth-death Markov process. The state transition rates are estimated by an iterative procedure. The procedure is developed according to the operation of the load-balancing policy. The proposed method is used to analyze the performance of a dynamic load-balancing policy, namely, symmetric policy. Extensive simulations are performed to study the accuracy of the method. This method provides accurate performance estimates for the symmetric policy for systems of various sizes when the mean job transfer delay is small compared to the average job service time.

An Analytical Model for Load Balancing on Symmetrical Multiprocessor Systems

In a distributed computing system, it is desirable to balance the work load among processors while keeping the communication overhead at a minimum. The nearest neighbor balancing strategy requires little communication overhead compared to the sophisticated dynamic load balancing policies. The question is whether this nearest neighbor balancing policy can achieve system balance. This paper investigates a typical nearest neighbor balancing strategy, called LAL (Local Average Load), in which the work load of a processor is averaged among its nearest neighbors at discrete time steps. The underlying systems considered are multiprocessor systems interconnected by generalized hypercube (GHC), torus, and other symmetric structures. Analytical models are developed for the load balancing scheme on the considered architectures, assuming that the amount of computation tasks arriving at or finished by a processor at each time step can be described by a random variable with some general distribution. We show that the expected difference between the actual load of a processor and the average load of the system is 0 and the variance of this difference is bounded by a constant independent of time. For each considered architecture, we derive a simple closed-form formula for the variance upper bound as a function of a system size and dimenslon. Furthermore, we generalize our analytical results to any linear symmetric load balancing scheme on symmetric distributed architectures.

A dynamic load-balancing policy with a central job dispatcher (LBC)

A dynamic load-balancing policy is proposed with a central job dispatcher called the LBC policy for distributed systems. The design of this policy is motivated by the operation of a single-queue multiserver queueing system, and the average job response time is the same as that of a single-queue multiserver system, which is the best achievable performance when the communication delay is reduced to zero. Hence, near-minimum average job response time is expected for distributed systems with high-speed communication subnets. The performance is studied for systems with nonnegligible job transfer delays in the following three aspects: average job response time, overhead due to information exchanges, and sensitivity to heterogeneous load. >