Load Balancing Policies Research Articles

OCReM: OpenStack-based cloud datacentre resource monitoring and management scheme

Managing virtualised computing, network and storage resources at large-scale in both public and private cloud datacentres is a challenging task. As an open source cloud operating system, OpenStack needs to be enhanced for managing cloud datacentre resources. In order to improve OpenStack functions to support cloud datacentre resource management, we present OCReM: OpenStack-based cloud datacentre resource monitoring and management scheme. First, we designed a virtual machine group life-cycle management module. Then, we designed and developed a cloud resource monitoring module based on the Nagios monitoring software and Libvirt interface. We conducted an integrated experiment to verify the performance improvement of group-oriented auto scaling and elastic load balancing policy based on real-time resource monitoring data. After that, we implemented the OCReM-EC2 hybrid cloud monitoring and auto scaling model. Finally, we analysed the prospective research direction and propose our future work.

High-Performance Centralized Content Delivery Infrastructure: Models and Asymptotics

We consider a centralized content delivery infrastructure where a large number of storage-intensive files are replicated across several collocated servers. To achieve scalable mean delays in file downloads under stochastic loads, we allow multiple servers to work together as a pooled resource to meet individual download requests. In such systems, basic questions include: How and where to replicate files? What is the impact of dynamic service allocation across request types, and whether such allocations can provide substantial gains over simpler load balancing policies? What are tradeoffs among performance, reliability and recovery costs, and energy? This paper provides a simple performance model for large systems towards addressing these basic questions.

Investigation on MDP-based radio access technology selection in heterogeneous wireless networks

The new generation of wireless networks is characterized by heterogeneity i.e. the coexistence of two or more radio access technologies (RAT) in the same geographical area. While this coexistence of RATs offers various advantages, it also imposes many challenges for the network operator, whose aim is to maximize the generated revenue while satisfying the customers’ increasing demands. One important mechanism in heterogeneous wireless networks (HWN) is the RAT selection. It is normally triggered when a new call arrives, and provides the decision on whether the call can be admitted or not, and by which RAT it has to be served. Different approaches can be used to tackle the problem of RAT selection in HWNs. In this paper, we study Markov Decision Process (MDP)-based RAT selection in a cellular/WLAN heterogeneous network where the maximization of the revenue is considered as objective. An optimal RAT selection policy is therefore derived. The performance of the optimal scheme is evaluated in comparison with two other policies, namely Cellular-First policy and Load Balancing policy.

Two Parallel Queues with Infinite Servers and Join the Shortest Queue Discipline

We study the stationary distribution of a system of two parallel M/M/∞ queues managed by the Join the Shortest Queue load balancing policy; one motivation for characterizing the efficiency of that policy is its potential application to resource allocation issues in cloud computing. For the general system with distinct service rates, we first show that the tail of each marginal queue length distribution exhibits a much faster decay than that of a Poisson distribution. Second, the determination of the joint stationary distribution is shown to reduce to the resolution of a pair of linear integral equations. In the case when service rates are identical (“symmetric case”), that pair of integral equations simplifies to a single Fredholm integral equation of the first kind whose solution is explicitly given in terms of Legendre polynomials; this enables us to entirely determine the stationary distribution of the system. We provide, in particular, asymptotics for the second moment and the tail of the queue length distribution.

CloudAnalyst : A Survey of Load Balancing Policies

ABSTRACT Cloud computing is one of the incredible technology which enable the new vision for IT industry. Nowadays, it has become a strong alternative for startup large as well as small scale organizations that only use the resources which actually required based on pay as per use. As Cloud Computing is growing continuously and clients from different parts of the world are demanding for the various services and better outcomes, the load balancing has become the challenge for the cloud provider. To accurately manage the available resources of the different cloud provider, resources have to be properly selected according to the properties of task. Many algorithms have been proposed to provide efficient mechanisms and assigning the client’s requests to available cloud nodes and aim to enhance the overall performance of the cloud and provide more satisfaction to user and efficient services. Initially this paper gives an introduction to cloud computing and load balancing. A detailed survey on different load balancing policy in cloud analyst, their advantages and drawback with obtainable solution and learn how to add new policy or customize existing load balancing policy.

Analysis of Service Broker Policies in Cloud Analyst Framework

The aim of the paper is to analyze different service broker policies that can be used with cloud and are been used in cloud analyst. With the advancement in the cloud computing technologies cloud is becoming more popular these days. With the increase in the number of clients and servers there is also arising needs of service broker. Broker is been used to assign a server to the client on clients request. Here I will first discuss about cloud computing, and then cloud analyst and the different broker policies embedded in it under round robin load balancing policy. Then with the help of simulation results will propose the best broker policy on the basis of performance.

An Enhancement in Service Broker Policy for Cloud-Analyst

Cloud computing means storing and accessing data and programs over the Internet instead of one’s personal computer's hard drive. The cloud is an analogy for the Internet. Now a day, it becomes very popular due to new trend and technology. It deals with large amount of data, so, it is necessary to simulate the behavior of cloud in real time environment. For this purpose, the simulation tool such as, cloud-analyst, cloudSim, are commonly used, which has been provided by laboratory CLOUD. The simulator cloud analyst uses different load balancing policy, service broker strategy with different parameters and has used as per requirement. This paper presents an improvement in service broker strategy, which enhances the performance of data center.

Virtual Machine Allocation Policy in Cloud Computing Using CloudSim in Java

Cloud computing is a very powerful concept that can be used to enhance the next generation data center and allow service provider to use data center capability provided by cloud and develop the application based on user requirement. Data center of this cloud computing has huge number of resources and list of applications (with different architecture, configuration and requirement for deployment) want to use those resource. Cloud computing environment uses virtualization concept and provides resources to application by creating and allocating virtual machine to specific application. There for resource allocation policies and load balance policies play very vital role in allocating and managing the resources among various application in clod computing life cycle. CloudSim is an extensible simulation toolkit that enables modeling and simulation of Cloud computing environments. The model proposed by this paper for dynamic load balance policy with considering different attributes and different service level agreements in cloud computing environment helps this environment to utilize their resources and improves performance. The proposed model uses Hungarian algorithm and the result is verified by simulating this model using CloudSim.

ORT Broker Policy: Reduce Cost and Response Time Using Throttled Load Balancing Algorithm

Abstract Cloud computing is simply define as “cloud” is delivery of demand services resources everything from application to data centers over the internet on a pay for use basis. In cloud computing load balancing is a technique to distribute the workload for balancing between two or more cloud server. Load balancing technique is defined some cases which are defining as guarantee continuity of the service provided manage, high traffic levels and always be ready for sudden peaks in traffic. Load balancing aims to optimize resource use, maintain the cost of data center and virtual machines, maximize throughput, minimize response time and avoid overload of any single resource. The main objective of this research paper is to reduce cost and response times using throttled load balancing policy across VM's in multi data center and optimise response time service broker policy. This study has evaluated throttled load balancing algorithm and their scheduling criteria like overall response time, Data center processing time and total cost of Virtual Machine and data transfer cost.

Cloud-aware data intensive workflow scheduling on volunteer computing systems

Volunteer computing systems offer high computing power to the scientific communities to run large data intensive scientific workflows. However, these computing environments provide the best effort infrastructure to execute high performance jobs. This work aims to schedule scientific and data intensive workflows on hybrid of the volunteer computing system and Cloud resources to enhance the utilization of these environments and increase the percentage of workflow that meets the deadline. The proposed workflow scheduling system partitions a workflow into sub-workflows to minimize data dependencies among the sub-workflows. Then these sub-workflows are scheduled to distribute on volunteer resources according to the proximity of resources and the load balancing policy. The execution time of each sub-workflow on the selected volunteer resources is estimated in this phase. If any of the sub-workflows misses the sub-deadline due to the large waiting time, we consider re-scheduling of this sub-workflow into the public Cloud resources. This re-scheduling improves the system performance by increasing the percentage of workflows that meet the deadline. The proposed Cloud-aware data intensive scheduling algorithm increases the percentage of workflow that meet the deadline with a factor of 75% in average with respect to the execution of workflows on the volunteer resources.

Comparative Study of Load Balancing Algorithms in Cloud Environment using Cloud Analyst

This is age of internet. There is hardly any domain where internet do not play significant role. Cloud Computing relates to internet computing that has revolutionized the whole world. Cloud Computing has undeniably benefited the users in many ways. Load Balancing is emerging as the hottest issue in cloud environment at present. Load due to client requests must be balanced by resources of service provider. In this study we proposed a theoretical framework which is combination of load balancing policy and service broker policy for cloud environment. This paper includes the comparison of the results of proposed algorithm with existing algorithm .This study accomplish that there is extreme reduction in response time at client part and reduction in datacenter request servicing time and resources of resource providers are managed efficiently and effectively by using the proposed algorithm. Several parameters are also identified and these are utilized to compare the traditional round robin load balancing policy with proposed load balancing strategy by using cloud analyst tool as simulator.

Load balancing for cluster systems under heavy-tailed and temporal dependent workloads

Large-scaled cluster systems have been employed in various areas by offering pools of fundamental resources. Efficient allocation of the shared resources in a cluster system is a critical but challenging issue, which has been extensively studied in the past few years. Despite the fact that existing load balancing policies, such as Random, Join Shortest Queue and size-based polices, are widely implemented in actual systems due to their simplicity and efficiency, the performance benefits of these policies diminish when workloads are highly variable and temporally correlated. In this paper, we propose a new load balancing policy, named ADuS, which attempts to partition jobs according to their present sizes and further rank the servers based on their loads. By dispatching jobs of similar sizes to the corresponding ranked servers, ADuS can adaptively balance user traffic and system load in a cluster and thus achieve significant performance benefits. Extensive trace-driven simulations using both synthetic and real traces show the effectiveness and robustness of ADuS under many different environments.

A greedy algorithm for load balancing jobs with deadlines in a distributed network

One of the most challenging issues when dealing with distributed networks is the efficiency of jobs load balancing. This paper presents a novel algorithm for load balancing jobs that have a given deadline in a distributed network assuming central coordination. The algorithm uses a greedy strategy for global and local decision making: schedule a job as late as possible. It has an increased overhead over other well-known methods, but the load balancing policy provides a better fit for jobs.

Advances in Multi-GPU Smoothed Particle Hydrodynamics Simulations

We present a multi-GPU version of GPUSPH, a CUDA implementation of fluid-dynamics models based on the smoothed particle hydrodynamics (SPH) numerical method. The SPH is a well-known Lagrangian model for the simulation of free-surface fluid flows; it exposes a high degree of parallelism and has already been successfully ported to GPU. We extend the GPU-based simulator to run simulations on multiple GPUs simultaneously, to obtain a gain in speed and overcome the memory limitations of using a single device. The computational domain is spatially split with minimal overlapping and shared volume slices are updated at every iteration of the simulation. Data transfers are asynchronous with computations, thus completely covering the overhead introduced by slice exchange. A simple yet effective load balancing policy preserves the performance in case of unbalanced simulations due to asymmetric fluid topologies. The obtained speedup factor (up to 4.5x for 6 GPUs) closely follows the expected one (5x for 6 GPUs) and it is possible to run simulations with a higher number of particles than would fit on a single device. We use the Karp-Flatt metric to formally estimate the overall efficiency of the parallelization.

Round Robin with Server Affinity: A VM Load Balancing Algorithm for Cloud Based Infrastructure

Cloud computing is an evolving computing paradigm that has influenced every other entity in the globalized industry, whether it is in the public sector or the private sector. Considering the growing importance of cloud, finding new ways to improve cloud services is an area of concern and research focus. The limitation of the available Virtual Machine Load balancing policies for cloud is that they do not save the state of the previous allocation of a virtual machine to a request from a Userbase and the algorithm requires execution each time a new request for Virtual Machine allocation is received from the Userbase. This problem can be resolved by developing an efficient virtual machine load balancing algorithm for the cloud and by doing a comparative analysis of the proposed algorithm with the existing algorithms. Keywords—Virtual Machine (VM), Server affinity, VM load balancer, CloudAnalyst, Data center, Cloudlet

A Hybird Virtual Machine Placement Aglrithm for Virtualized Desktop Infrastructure

As we all kown, The virtual machine placement is one kind of bin-packing problem. By optimizing placement of virtual machine. We can improve VM performance, enhance resource utilization, reduce energy comsumption. After analysis the existing virtual machine placement aglrithm. We propose a hybird virtual machine placement aglrithm (HTA) which based on network latency threshold for the requirement of low network latence and low VM migraiton ratio in Virtualized Desktop Infrastructure. It elect qualified node set based on network latency threshold and palce the virtual machines with load-balance policy, taking into account the preformance of the network and vitual machines. According to analysis and comparison. The simulation result show that the algorithm can effectively lessen the network latency and reduce the VM migration ratio.

Performance Evaluation of Large Scaled Applications using Different Load Balancing Tactics in Cloud Computing

There is no doubt that workload distribution in cloud computing is one of the most important factors that regulates and dictates, directly or indirectly, its behavior and affects system performance. But geographical location of the datacenter and the user group is also an important factor that affects the overall performance of the system. This paper represents the performance analysis of three load balancing policies in combination with different broker policies for large-scaled applications using different infrastructural environments. The study has been made using CloudAnalyst: A CloudSim-based tool for modeling and analysis of large scale cloud computing environments. Experimental results reveal that large scaled software systems like FACEBOOK, TWITTER, ORKUT and e-commerce applications etc. can minimize their costs and improve service quality to the end users by making a right choice of service provider in the cloud market. The cloud service providers can also optimize their processing time by applying a best resource provisioning policy.

一种动态的客户端负载均衡机制

Web applications traffic demand fluctuates widely and usually unpredictably. The common practice of provisioning a fixed capacity would either result in unsatisfied customers (when the resources allocated to a specific customer are too low) or waste valuable capital investment (when the resources allocated to a specific customer are too much). By leveraging an infrastructure clouds on-demand, pay-per-use capabilities, we finally can match the systems capacity with the demand in real time. Although the static client-side load balancing mechanism has been used by many clusters, it does not suit the cluster which needs to change nodes and load balancing policies in runtime. This paper proposes a dynamic client-side load balancing mechanism. By integrating technologies such as distributed cluster-view maintenance and control flow, this mechanism not only has good scalability but also introduces many dynamic features into the client-side load balancing, providing effective support for the nodes dynamic join and exit as well as the adaptive adjustment of the load balancing policies. Since the framework has been implemented in an open-source JEE application server named PKUAS, this paper also describes some key implementation features and analyzes the evaluation results of the experiments to show the effectiveness of the mechanism from different aspects.

LatLong: Diagnosing Wide-Area Latency Changes for CDNs

Minimizing user-perceived latency is crucial for Content Distribution Networks (CDNs) hosting interactive services. Latency may increase for many reasons, such as interdomain routing changes and the CDN's own load-balancing policies. CDNs need greater visibility into the causes of latency increases, so they can adapt by directing traffic to different servers or paths. In this paper, we propose a tool for CDNs to diagnose large latency increases, based on passive measurements of performance, traffic, and routing. Separating the many causes from the effects is challenging. We propose a decision tree for classifying latency changes, and determine how to distinguish traffic shifts from increases in latency for existing servers, routers, and paths. Another challenge is that network operators group related clients to reduce measurement and control overhead, but the clients in a region may use multiple servers and paths during a measurement interval. We propose metrics that quantify the latency contributions across sets of servers and routers. Based on the design, we implement the LatLong tool for diagnosing large latency increases for CDN. We use LatLong to analyze a month of data from Google's CDN, and find that nearly 1% of the daily latency changes increase delay by more than 100 msec. Note that the latency increase of 100 msec is significant, since these are daily averages over groups of clients, and we only focus on latency-sensitive traffic for our study. More than 40% of these increases coincide with interdomain routing changes, and more than one-third involve a shift in traffic to different servers. This is the first work to diagnose latency problems in a large, operational CDN from purely passive measurements. Through case studies of individual events, we identify research challenges for managing wide-area latency for CDNs.

A Hierarchical Load Balancing Policy for Grid Computing Environment

With the rapid development of high-speed wide-area networks and powerful yet low-cost computational resources, grid computing has emerged as an attractive computing paradigm. It provides resources for solving large scientific applications. It is typically composed of heterogeneous resources such as clusters or sites at different administrative domains connected by networks with widely varying performance characteristics. The service level of the grid software infrastructure provides two essential functions for workload and resource management. To efficiently utilize the resources at these environments, effective load balancing and resource management policies are fundamentally important. This paper addresses the problem of load balancing and task migration in grid computing environments. We propose a fully decentralized two-level load balancing policy for computationally intensive tasks on a heterogeneous multi-cluster grid environment. It resolves the single point of failure problem which many of the current policies suffer from. In this policy, any site manager receives two kinds of tasks namely, remote tasks arriving from its associated local grid manager, and local tasks submitted directly to the site manager by local users in its domain, which makes this policy closer to reality and distinguishes it from any other similar policy. It distributes the grid workload based on the resources occupation ratio and the communication cost. The grid overall mean task response time is considered as the main performance metric that need to be minimized. The simulation results show that the proposed load balancing policy improves the grid overall mean task response time.