Adaptive Load Balancing Scheme Research Articles

An adaptive load balancing strategy for stateful join operator in skewed data stream environments

As one of the most computationally intensive operations in stream processing applications, join operation can cause severe load imbalance problem when dealing with skewed data. Most of the popular solutions focused on monitoring-based dynamic balancing strategies, making it difficult to quickly adapt to the changing frequency of data stream, and sometimes failing the balancing strategies that try to address the skewed load in the cluster. To address these issues, we propose to use the prediction results of a deep reinforcement learning model and adjust the grouping strategy in advance before the frequency change of data stream. It will enable the system to quickly adapt to data stream fluctuation, while managing the resources for effective resource utilization. The following contributions are made in this paper: 1) Explore the main factors that trigger the load skewness problem in distributed stream join systems and carefully model the load balancing problem at the application level. 2) Develop a Gated Recurrent Unit Sequence to Sequence model to predict key frequency distribution of streams, and propose a dynamic grouping algorithm and a feedback-based resource elasticity scaling algorithm to solve the load imbalance problem caused by hot keys in real time. 3) Design and implement an adaptive stream join system Aj-Stream based on the prediction model and the proposed algorithm on Apache Storm. 4) Evaluate the system performance through extensive experiments on a large scale real-world dataset and multiple synthetic datasets. The experimental results demonstrate that the Aj-Stream proposed in this paper exhibits stable throughput and latency performance with both static data streams of varying skewnesses and dynamic data streams. In comparison to existing stream-connected systems, Aj-Stream demonstrated a 22.1% increase in system throughput and a 45.5% decrease in system latency when dealing with frequently fluctuating data streams.

Load Balanced Data Transmission Strategy Based on Cloud–Edge–End Collaboration in the Internet of Things

To improve the response speed and quality of Internet of Things (IoT) services and reduce system operating costs, this paper refines the edge layer according to the different data transmission capabilities of different edge devices, constructs a four-layer heterogeneous IoT framework under cloud–edge–end (CEE) collaboration, and gives the corresponding data hierarchical transmission strategy, so as to effectively process sensitive data such as real-time, near-real-time, and non-real-time data. Meanwhile, the link based high-performance adaptive load balancing scheme is developed to achieve the dynamic optimal allocation of system resources. Simulation results demonstrate that the data hierarchical transmission strategy based on a CEE collaboration framework can not only make full use of resources and improve the successful delivery rate of packets but can also greatly reduce the end-to-end transmission delay of data. Especially, compared with the cloud-mist framework without refining the edge layer, the data transmission rate based on CEE collaboration architecture is increased by about 27.3%, 12.7%, and 8%, respectively, in three network environments of light-weight, medium, and heavy load.

An adaptive load balancing scheme for software-defined network controllers

Software-defined networking (SDN) is considered as one of the promising paradigms for the future Internet. It allows users to configure the network depending on the application-level requirements. One of the major challenges in SDN is the uneven distribution of traffic load in the controllers that degrades system performance. In this paper, we propose a self-adaptive load balancing (SALB) scheme that balances load among multiple controllers dynamically with multiple switch migration from source controllers to target controllers. The key feature of our scheme is an effective distribution of load under high load condition while considering the distance between switches and target controllers simultaneously. The efficacy of our proposed scheme is demonstrated through experimentation, where we compare our scheme with existing approaches. The results show that SALB experiences a small number of packet drops, which is less than 1.23% of the total number of message exchanges among the controllers.

Adaptive Load Balancing Strategy Based on LVS

Linux Virtual Server (LVS) is a load balancing server that deployed on a cluster of real servers and the load balancer running on the Linux operating system. In the scheduling module, we designed an adaptive load balancing strategy. Firstly, we introduce the traffic state prediction. The algorithm we choose is Radial Basis Function (RBF) neural network. Secondly, we update weights according to the real-time usage of CPU, memory and hard disk. Then we conduct a series of experiments and the adaptive strategies have a better performance on load balancing.

Adaptive Load Balancing Scheme For Data Center Networks Using Software Defined Network

A new adaptive load balancing scheme for data center networks is proposed in this paper by utilizing the characteristics of Software Defined Networks. Mininet was utilized for the purpose of emulating and evaluating the proposed design, Miniedit was utilized as a GUI tool. In order to obtain a similar environment to the data center network, Fat-Tree topology was utilized. Different scenarios and traffic distributions were applied in order to cover as much cases of the real traffic as possible. The suggested design showed superiority over the traditional scheme in term of throughput and loss rate for all the evaluated scenarios. Two scenarios were implemented; the proposed algorithm presented a loss-free performance compared to 15% to 31% loss rate in the traditional scheme for the first scenario. The proposed scheme showed up to 81% improvement in the loss rate in the second scenario. In term of throughput, the proposed scheme maintained the same level of throughput in the first scenario compared to an average of 5Mbps reduction in the throughput when using the traditional scheme. While in the second scenario, the new scheme outperformed the traditional scheme by showing an improvement of up to 16.6% in the throughput value.

TRINI: an adaptive load balancing strategy based on garbage collection for clustered Java systems

SummaryNowadays, clustered environments are commonly used in high‐performance computing and enterprise‐level applications to achieve faster response time and higher throughput than single machine environments. Nevertheless, how to effectively manage the workloads in these clusters has become a new challenge. As a load balancer is typically used to distribute the workload among the cluster's nodes, multiple research efforts have concentrated on enhancing the capabilities of load balancers. Our previous work presented a novel adaptive load balancing strategy (TRINI) that improves the performance of a clustered Java system by avoiding the performance impacts of major garbage collection, which is an important cause of performance degradation in Java. The aim of this paper is to strengthen the validation of TRINI by extending its experimental evaluation in terms of generality, scalability and reliability. Our results have shown that TRINI can achieve significant performance improvements, as well as a consistent behaviour, when it is applied to a set of commonly used load balancing algorithms, demonstrating its generality. TRINI also proved to be scalable across different cluster sizes, as its performance improvements did not noticeably degrade when increasing the cluster size. Finally, TRINI exhibited reliable behaviour over extended time periods, introducing only a small overhead to the cluster in such conditions. These results offer practitioners a valuable reference regarding the benefits that a load balancing strategy, based on garbage collection, can bring to a clustered Java system. Copyright © 2016 John Wiley & Sons, Ltd.

Performance of Replica-exchange Wang-landau Sampling for the 2D Ising Model: A Brief Survey

Abstract We report a brief performance study of the replica-exchange Wang-Landau algorithm, a recently proposed parallel realization of Wang-Landau sampling, using the 2D Ising model as a test case. The simulation time is found to scale inversely with the square root of the number of subwindows (and thus number of processors) used to span the global parameter space. We also investigate the “time profiles” for random walkers in different subwindows to complete iterations, which will aid the development of an adaptive load-balancing scheme.

A GPU/CPU Programming Model for CFD Simulation

In recent years, the highly parallel graphics processing unit (GPU) is rapidly gaining maturity as a powerful engine for high performance computer. More and more researchers try to port the computational fluid dynamics (CFD) simulations into heterogeneous computers. However, most researchers focus on exploring the computational capability of GPU, while ignore the computational capability of CPU. In order to utilize the computational capability of CPU and GPU, we propose a hybrid CUDA/OpenMP parallel programming model. And we proposed an adaptive load balancing scheme to distribute the workload among CPUs and GPUs. With this programming model, we implement a high-order CFD program on “Tianhe-1A” supercomputer system. The performance results validate the workload distribution scheme.

Task Partitioning and Load Balancing Strategy for Matrix Applications on Distributed System

In this paper, we present a load-balancing strategy (Adaptive Load Balancing strategy) for data parallel applications to balance the work load effectively on a distributed system. We study its impact on computation-hungry matrix multiplication application. The ALB strategy enhances the performance with features such as intelligent node selection, pre-task assignment, adaptive task sizing and buffer allocation, and load balancing. The ALB strategy exhibits reduced nodes idle time and inter process communication time, and improved speed up as compared to Run Time task Scheduling strategy.

A dynamic and adaptive load balancing strategy for parallel file system with large-scale I/O servers

Many solutions have been proposed to tackle the load imbalance issue of parallel file systems. However, all these solutions either adopt centralized algorithms, or lack considerations for both the network transmission and the tradeoff between benefits and side-effects of each dynamic file migration. Therefore, existing solutions will be prohibitively inefficient in large-scale parallel file systems. To address this problem, this paper presents SALB, a dynamic and adaptive load balancing algorithm which is totally based on a distributed architecture. To be also aware of the network transmission, SALB on the one hand adopts an adaptively adjusted load collection threshold in order to reduce the message exchanges for load collection, and on the other hand it employs an on-line load prediction model with a view to reducing the decision delay caused by the network transmission latency. Moreover, SALB employs an optimization model for selecting the migration candidates so as to balance the benefits and the side-effects of each dynamic file migration. Extensive experiments are conducted to prove the effectiveness of SALB. The results show that SALB achieves an optimal performance not only on the mean response time but also on the resource utilization among the schemes for comparison. The simulation results also indicate that SALB is able to deliver high scalability.

An Adaptive Task-Core Ratio Load Balancing Strategy for Multi-core Processors

With the proliferation of multi-core processors in servers, desktops, game consoles, mobile phones and a magnitude of other embedded devices; the need to ensure effective utilization of the processing cores becomes essential. This calls for research and development emphasis for a well engineered operating systems load balancer for these multi-core processors. In this paper, an adaptive load balancing strategy is presented. The adaptive load balancer will trigger tasks migration based on the tasks to processing core ratio, as well as when a processing core becomes idle. In our work, we utilize LinSched, a Linux operating system scheduler simulator, to analyze the number of task migrations. The Linux operating system is representative of the whole spectrum of computing as it is used in supercomputers, servers, desktops, mobile phones and embedded devices. Results from the simulation show that unnecessary task migrations were eliminated whilst maintaining the load balancing function effectively, as compared to the default strategy employed by the Linux operating system. The overheads introduced by the adaptive load balancer were measure through implementing it in a Linux kernel and measurements were made using the hackbench scalability test. The implementation proves to have negligible effect on the scalability and we can conclude that it does not introduce overheads. From our research, it shows that the adaptive load balancer provides a scalable solution for a lower and more consistent triggering of task migrations.

An adaptive load balancing management technique for RFID middleware systems

AbstractIn recent years, Radio Frequency Identification (RFID) industries have taken a great interest in utilizing the benefits of RFID for supply chain management, inventory control and various other applications. This paper proposed an adaptive load balancing technique for RFID middleware systems to meet the demands of scalability and heterogeneity. First, we explored five basic load balancing policies, namely, information policy, job selection policy, transfer policy, initiation policy and location policy. Eighteen load balancing schemes were then proposed for RFID middleware systems that were combinations of various types of the five basic load balancing policies. Our empirical study suggested that these load balancing strategies performed differently under different workload statuses. Finally, an adaptive load balancing strategy was proposed. The load balancing schemes and the proposed adaptive load balancing strategy have been implemented in the RFID Middleware Load Management System (RM‐LMS). Copyright © 2010 John Wiley & Sons, Ltd.

SALSA: QoS-aware load balancing for autonomous service brokering

The evolution towards “Software as a Service”, facilitated by various web service technologies, has led to applications composed of a number of service building blocks. These applications are dynamically composed by web service brokers, but rely critically on proper functioning of each of the composing subparts which is not entirely under control of the applications themselves. The problem at hand for the provider of the service is to guarantee non-functional requirements such as service access and performance to each customer. To this end, the service provider typically divides the load of incoming service requests across the available server infrastructure. In this paper we describe an adaptive load balancing strategy called SALSA (Simulated Annealing Load Spreading Algorithm), which is able to guarantee for different customer priorities, such as default and premium customers, that the services are handled in a given time and this without the need to adapt the servers executing the service logic themselves. It will be shown that by using SALSA, web service brokers are able to autonomously meet SLAs, without a priori over-dimensioning resources. This will be done by taking into account a real time view of the requests by measuring the Poisson arrival rates at that moment and selectively drop some requests from default customers. This way the web servers’ load is reduced in order to guarantee the service time for premium customers and provide best effort to default customers. We compared the results of SALSA with weighted round-robin (WRR), nowadays the most used load balancing strategy, and it was shown that the SALSA algorithm requires slightly more processing than WRR but is able to offer guarantees – contrary to WRR – by dynamically adapting its load balancing strategy.

An Adaptive Threshold Load Balancing Scheme for the End-to-End Reconfigurable System

In this paper, an adaptive threshold load balancing (ATLB) scheme is proposed for the joint load control (JOLDC) in the end-to-end reconfigurable system. The dynamic threshold based on heterogeneous load measurement is designed to adapt to varying load conditions for efficient resource utilization, while the network-initiated vertical handover (NIVHO) process is adopted for the timely control by exploiting the network-controlled reconfiguration management. Furthermore, considering terminal reconfigurability and the service variety, a multi-objective terminal selection algorithm is designed for NIVHO to minimize the system overhead and the impact on end-users. Simulation results show that the proposed scheme is effective in relieving congestion and improving system performance across the overlapped heterogeneous radio networks as well as the users' satisfaction.

Adaptive load balancing scheme in ad hoc networks

An adaptive load balancing scheme is proposed to balance the load in ad hoc networks. The new scheme can be applied in most on-demand routing protocols resulting in significant performance improvement. The proposed scheme is applied to the ad hoc on-demand distance vector (AODV) routing protocol. Simulation results show that the network load is balanced on the whole, and performance in packet loss rate, routing overhead and average end-to-end delay is also improved.

An Adaptive Hybrid Task Partitioning, Scheduling (AHTS), and Load Balancing Strategy for Heterogeneous Distributed System

Heterogeneous Parallel Distributed Processing (PDP) systems, which exploit the aggregate power of a network of workstations (WSs) and Personal Computers (PCs) are an inexpensive alternative to the dedicated parallel supercomputing systems. As these systems are widely available in academic and industrial environments, it is becoming popular to use these computing resources to solve time consuming applications. The main problem with such type of cluster computing environment is the continuous change in the performance of individual WSs/PCs which requires an efficient task partitioning, scheduling, and load balancing to get better performance.In this paper, we investigate the problem of static and runtime tasks allocation for parallel-distributed image computing system. The PDP system has heterogeneity in processors and resources. For such system, we propose an Adaptive Hybrid Task partitioning, Scheduling (AHTS), and load balancing strategy. The investigation is examined on a manager/master and workers model of PDP system. The measured results show that, the AHTS strategy dramatically improves the performance of PDP image computing system and remedy the defects of static and Runtime Task Scheduling (RTS) strategies.

Dynamic traffic assignment on parallel computers in TRANSIMS

We describe part of the framework of the TRANSIMS traffic research project at the Los Alamos National Laboratory. It includes parallel implementations of a route planner and of a microscopic traffic simulation, and iterative re-planning for dynamic route assignment. The parallel implementation uses domain decomposition and an adaptive load-balancing scheme. We present performance figures for street networks with up to 20 000 links, and estimates for much larger networks, as will be used for the next TRANSIMS case study.

Integrating Parallelization Strategies for Linkage Analysis

We describe a second parallel implementation of the ILINK program from the LINKAGE package that improves on our previous implementation. To improve running time we integrated the strategy of parallel estimation of the gradient at a candidate recombination fraction vector with a previously implemented strategy for evaluating in parallel the likelihood at one vector. We also integrated an adaptive loadbalancing strategy in conjunction with our previous static loadbalancing strategy. We implemented a strategy for partitioning input pedigrees, but this slowed down the program; we give some evidence for what the problems are. To best exploit parallelism at all levels of the program and to take advantage of both coarse-grain and fine-grain parallelism it is necessary to combine multiple algorithmic strategies.

Adaptive load balancing with bidirectional piggybacking

AbstractComputer network technology is no longer experimental. Having come of age, this technology is now both practical and necessary. Downsizing of computers and open and high‐speed networks have required multimedia processing as well as traditional data processing on a networked system connected by various types of computers. However, in current network systems, users are responsible for job submission, while simultaneously considering what and where the necessary and available resources are. This will create a heavier burden in the future. A framework of load balancing with distributed dispatchers that uses an analysis of network delay and job characteristics is proposed here. Examining an adaptive load balancing scheme with bidirectional piggybacking permits basic study of this framework. First a bidirectional piggybacking is proposed that uses the results of a job as an information carrier for a remote server and that uses the request of a job as an information carrier for a local server. We also propose a load balancing scheme using bidirectional piggybacking. This scheme is then compared with other schemes on homogeneous and heterogeneous environments with and without communication delay. Finally, the reliability of piggybacked information on the arrival rate of local/remote job, communication delay, etc. are discussed. The effects of this scheme based on bidirectional piggybacking are also discussed.