Load Balancing Features Research Articles

Load Balancing in Cloud Computing Using Meta Heuristic Algorithm

Cloud computing involves load balancing, a technique for evenly distributing work across a cloud-based environment. Enterprises authorized applications to be managed, with these apps typically developed by the network's provider. The cloud system relies on multiple computers, supporting stakeholders in utilizing it for load balancing. It ensures virtual machines remain balanced, effectively distributing the workload among them and helping companies handle network traffic and workload issues. The load balancing system reduces imbalances among virtual machines, using the Bat algorithm to encrypt essential load balancing features. Our paper presents these techniques as a meta-heuristic algorithm. Keywords— Cloud Computing, Load Balancing, Virtual Machines, Workload.

Towards building a Deep Learning based Automated Indian Classical Music Tutor for the Masses

• Addresses a humanitarian concern and presents a possible social innovation. • State-of-the-art Deep Learning techniques for music classification. • Deployed in a highly scalable, orchestrated, and containerized cloud environment. • A substantial literature survey. • Results are acceptably accurate. Music can play an important role in the well-being of the world. Indian classical music is unique in its requirement for rigorous, disciplined, expert-led training that typically goes on for years before the learner can reach a reasonable level of performance. This keeps many, including the first author of this paper, away from mastering the skill. The problem is particularly compounded in rural areas, where the available expertise may be limited and prohibitively expensive, but the interest in learning classical music still prevails, nevertheless. Machine Learning has been complementing, enhancing, and replacing many white-collar jobs and we believe it can help with this problem as well. This paper describes efforts at using Machine Learning techniques, particularly, Long Short-Term Memory for building a system that is a step toward provisioning an Indian Classical Music Tutor for the masses. The system is deployed in the cloud using orchestrated containerization for potential worldwide access, load balancing, and other robust features.

Hybrid Optical-Electrical Data Center Networking: Challenges and Solutions for Bandwidth Resource Optimization

Fast optical switching (FOS) is the most promising technology for data center networks (DCNs) to meet data rate requirements beyond 800 Gb/s, considering its transparent modulation format, low power consumption and dense port integration. However, FOS, leveraging optical packet switching (OPS) technology, is hardly compatible with the electrical switches that are widely used in the current hybrid optical/electrical switching data center networks (Hoe-DCNs). The diversity of topologies and multi-switching granularity increases the risk of service disruption and reduces bandwidth efficiency. Based on the IP-over-OPS system, we propose bandwidth resource optimization techniques with the aid of hybrid polling contention resolution and microsecond-level load balancing in Hoe-DCNs. We first present an optical switching architecture to show its performance characteristics in the context of nanosecond-level FOSs in terms of resource management and setup optimization. Based on the fabricated FOS structure, we discuss the emerging problem of packet contention and introduce a bufferless scheduling solution. The characteristics of load balancing for Hoe-DCN are then investigated and evaluated by simulating Hoe-DCNs. Finally, the numerical results are given and analyzed based on a large-scale DCN testbed.

A Secure and Efficient Method to Protect Communications and Energy Consumption in IoT Wireless Sensor Networks

The rapid growth of technology has resulted in the deployment of a large number of interconnected devices, resulting in a wide range of new societal services. Wireless sensor networks (WSNs) are a promising technology which is faced with the challenges of operating a large number of sensor nodes, information gathering, data transmission, and providing a means to act in different scenarios such as monitoring, surveillance, forest fire detection, and many others from the civil to military spectrum. The deployment scenario, the nature of the sensor-equipped nodes, and their communication methods make this architecture extremely vulnerable to attacks, tampering, and manipulation than conventional networks. Therefore, an optimal solution to ensure security in such networks which captures the major constraints of the network in terms of energy utilization, secured data transmission, bandwidth, and memory fingerprint to process data is required. This work proposes a fast, reliable, and secure method of key distribution and management that can be used to ensure the integrity of wireless sensor networks’ communications. Moreover, with regards to efficient energy utilization, an improvement of the Low Energy Adaptive Clustering Hierarchy (LEACH) algorithm (a cluster routing protocol that is mainly used in WSN) has been proposed to enhance the networks’ energy efficiency, simplicity, and load-balancing features. Therefore, in this paper, we propose a combination of a distributed key exchange and management methods based on elliptic curve cryptography to ensure security of node communication and an improved routing protocol based on the LEACH protocol to demonstrate better performance in parameters such as network lifespan, dead nodes, and energy consumption.

GraphBLAST: A High-Performance Linear Algebra-based Graph Framework on the GPU

High-performance implementations of graph algorithms are challenging to implement on new parallel hardware such as GPUs because of three challenges: (1) the difficulty of coming up with graph building blocks, (2) load imbalance on parallel hardware, and (3) graph problems having low arithmetic intensity. To address some of these challenges, GraphBLAS is an innovative, on-going effort by the graph analytics community to propose building blocks based on sparse linear algebra, which allow graph algorithms to be expressed in a performant, succinct, composable, and portable manner. In this paper, we examine the performance challenges of a linear-algebra-based approach to building graph frameworks and describe new design principles for overcoming these bottlenecks. Among the new design principles is exploiting input sparsity , which allows users to write graph algorithms without specifying push and pull direction. Exploiting output sparsity allows users to tell the backend which values of the output in a single vectorized computation they do not want computed. Load-balancing is an important feature for balancing work amongst parallel workers. We describe the important load-balancing features for handling graphs with different characteristics. The design principles described in this paper have been implemented in “GraphBLAST”, the first high-performance linear algebra-based graph framework on NVIDIA GPUs that is open-source. The results show that on a single GPU, GraphBLAST has on average at least an order of magnitude speedup over previous GraphBLAS implementations SuiteSparse and GBTL, comparable performance to the fastest GPU hardwired primitives and shared-memory graph frameworks Ligra and Gunrock, and better performance than any other GPU graph framework, while offering a simpler and more concise programming model.

Network slicing with load-balancing for task offloading in vehicular edge computing

The support of edge computing for vehicular technologies gained increasing momentum with 5G to fulfill efficient offloading tasks from vehicles towards the edge nodes. Accordingly, vehicles demanding powerful computation and large storage resources will be directed to communicate with the nearest edge computing nodes hosted at a wireless 5G new generation nodes (gNBs) or a road side units (RSUs). To efficiently utilize the edge nodes' resources, network slicing and load-balancing features can greatly help in that, therefore, this paper proposes an algorithm for Vehicular Edge Computing (VEC) with network slicing and load-balancing based on resources utilization, denoted as VECSlic-LB, specifically dedicated for offloading tasks from vehicles to edge nodes at gNBs or RSUs. The algorithm can holistically view and manage the whole network, and use network function virtualization framework to manage the data plane. VECSlic-LB can handle a mix of slicing configurations, capable of balancing the loads between various slices per node, and can support multiple edge computing nodes. Several simulations were conducted comparing the performance of the proposed algorithm to the optimal solution, resulting on very close acceptance ratios as the optimal solution, and also was evaluated against a recent reference algorithm, providing more efficient resource utilizations ratios, saving up to 48% of the resources better than the state-of-art algorithm.

KINERJA PROTOKOL NETWORK REDUNDANSI VRRP, HSRP DAN GLBP PADA LAYANAN VIDEO STREAMING

Nowadays streaming video services have become the main need for internet users. Tokeep the streaming video service awake, an alternative path is needed to avoid network failure.In creating an alternative path, you can use first hop redundancy protocol (FHRP). In FHRPthere are three redundancy protocols including VRRP, HSRP and GLBP. The purpose of thisstudy is to find out better protocol performance, especially by using streaming video services.The method used is to simulate redundancy protocol using GNS-3 software with 3 scenariosthat run including normal network, down network and recovery network. As for testing eachscenario including testing routers with priority, route trace route, link transfer time and quality ofservice (QoS) video streaming. Based on the results of the study, it was obtained that GLBPprotocol has better performance with an average link transfer time of 1.51 seconds, delay of8.12 ms, packet loss of 0 % and throughput of 1,248 Mbps. This is because in GLBP protocolthere is a load balancing feature that can split the load equally and utilize resources moreeffectively from each router.

Addressless: A new internet server model to prevent network scanning.

Eliminating unnecessary exposure is a principle of server security. The huge IPv6 address space enhances security by making scanning infeasible, however, with recent advances of IPv6 scanning technologies, network scanning is again threatening server security. In this paper, we propose a new model named addressless server, which separates the server into an entrance module and a main service module, and assigns an IPv6 prefix instead of an IPv6 address to the main service module. The entrance module generates a legitimate IPv6 address under this prefix by encrypting the client address, so that the client can access the main server on a destination address that is different in each connection. In this way, the model provides isolation to the main server, prevents network scanning, and minimizes exposure. Moreover it provides a novel framework that supports flexible load balancing, high-availability, and other desirable features. The model is simple and does not require any modification to the client or the network. We implement a prototype and experiments show that our model can prevent the main server from being scanned at a slight performance cost.

Congestion Control in SDN-Based Networks via Multi-Task Deep Reinforcement Learning

Congestion control is a fundamental network task that modulates the data transmission rates of traffic sources to efficiently utilize network capacity. With the advent of machine learning, congestion control based on deep reinforcement learning is the subject of extensive attention. At present, research on machine-learning-based congestion control is mainly focused on single-task scenarios -- to design a control strategy that is only used to reduce congestion. This article studies the congestion control problem based on machine learning in multi-task scenarios. Specifically, we propose a congestion control model based on multi-task deep reinforcement learning. The model takes congestion control as the main task and load balancing as the auxiliary task. Compared to the single-task method, our model can better represent the network environment by learning the shared representation of congestion features and load balancing features. Moreover, network traffic control may involve both congestion control and load balancing. Therefore, learning multiple tasks jointly while exploiting commonalities and differences across tasks can help reduce the cost of task coordination. We use software defined networking to decouple the data and control planes, making network control strategies more flexible than traditional networks. To the best of our knowledge, this is the first time multi-task learning has been applied to network traffic control. Experimental results show that the method is efficient.

Optimizing the Topology and Energy-Aware VM Migration in Cloud Computing

The advancements in virtual machine migration (VMM) have been trending due to its effective load balancing features in cloud infrastructure. Previously, data centers were used for handling VMs organized in racks. These racks are arranged in a spanning tree topology with a high bandwidth. Thus, the cost for moving the data between servers is highest when the racks are far from each other. This work addresses this issue and proposed VMM strategy based on self-adaptive D-Crow algorithm (S-DCrow) that incorporates adaptive constants in Dragonfly-based Crow (D-Crow) optimization algorithm based on the proposed topology model. The proposed S-DCrow describes a migrating model, which is based on topology, energy consumption, load, and migration cost. Here, the network is organized in a spanning tree topology and is adapted by proposed S-DCrow for optimal VMM. The performance of the proposed S-DCrow shows superior performance in terms of load, energy consumption, and migration cost with the values of 0.1417, 0.1009, and 0.1220, respectively.

RAT Association for Autonomic IoT Systems

The next generation of wireless communications consists of heterogeneous networks that operate over different technologies and standards. Co-existence of multiple technologies aims to enhance the network capacity in order to satisfy the increasing number of connected devices. However, in order to meet high performance, it is important to assign the right device to the right RAT. In this article, we propose a novel distributed approach to solve the RAT selection problem, in a self-organized IoT context. Our scheme is mainly based on tools from MG theory. We show that a stable association between connected IoT devices and available RATs can be met while using a fully distributed algorithm that respects connected devices requirements and RAT constraints as well. The proposed MG assignment allows IoT devices, with limited energy budget, to improve their energy efficiency and to reduce the data transmission cost over the serving RAT at very light signaling overhead. Moreover, extensive simulations show that the MG based approach outperforms the conventional throughput maximization association in terms of energy efficiency and throughput. Our scheme also exhibits some nice load balancing features.

Uniform Minimal First: Latency Reduction in Throughput-Optimal Oblivious Routing for Mesh-Based Networks-on-Chip

Mesh-based networks-on-chips (NoCs) are increasingly used for on-chip communications in embedded multicore processors. O1TURN is a well-known oblivious routing algorithm for mesh-based NoCs that has been shown to be worst-case throughput optimal for even network radices, but not for odd radices. More recently, another oblivious routing algorithm called U2TURN has been shown to be worst-case throughput optimal for both odd and even radices. This is accomplished by load-balancing among 2-turn paths in XYX or YXY routing, including nonminimal paths. Besides being worst-case throughput optimal, U2TURN achieves higher throughput than O1TURN under adversarial traffic. However, for random traffic, where the traffic is inherently load-balanced, O1TURN achieves lower network latency since it only considers minimal paths. In this letter, we propose a hybrid oblivious routing algorithm called uniform-minimal-first (UMF) routing. UMF works by exploiting any inherent load-balancing characteristics of the traffic pattern to reduce packet latency, but it retains the throughput optimality of U2TURN for both odd and even radices, and it achieves the performance of U2TURN under adversarial traffic. UMF is very inexpensive to implement as it just requires incrementing two small counters, but only for the head flit when a node injects a new packet (not for any pass-through traffic). These simple updates can be performed one cycle ahead of packet injection to avoid slowing down any router pipeline stage. Despite its simplicity, UMF outperforms U2TURN by 23.7% under random traffic and O1TURN by 12.6% under adversarial traffic.

Performance evaluation and analysis of sparse matrix and graph kernels on heterogeneous processors

Heterogeneous processors integrate very distinct compute resources such as CPUs and GPUs into the same chip, thus can exploit the advantages and avoid disadvantages of those compute units. We in this work evaluate and analyze eight sparse matrix and graph kernels on an AMD CPU–GPU heterogeneous processor by using 956 sparse matrices. Five characteristics, i.e., load balancing, indirect addressing, memory reallocation, atomic operations, and dynamic characteristics are our major considerations. The experimental results show that although the CPU and GPU parts access the same DRAM, very different performance behaviors are observed. For example, though the GPU part in general outperforms the CPU part, it cannot achieve the best performance in all cases given by the CPU part. Moreover, the bandwidth utilization of atomic operations on heterogeneous processors can be much higher than a high-end discrete GPU.

Cost-Aware Stochastic Compressive Data Gathering for Wireless Sensor Networks

Data gathering is a crucial function of wireless sensor networks (WSNs). In resource-limited WSNs, it is critical to improve cost-efficiency and prolong network lifetime. Sensor networks utilizing deterministic routing paths are particularly vulnerable to attacks. Additionally, repeated use of the same path will introduce load unbalance. In this paper, we propose a cost-aware stochastic compressive data gathering for WSNs. In contrast to traditional compressive sensing (CS) based algorithms that implicitly assume uniform transmission cost, our proposed scheme will consider the cost diversity into the CS data gathering framework. The Markov chain-based model will be used to characterize the stochastic data gathering process. An optimization problem is formulated to minimize the total expected cost subjected to the constraints on global degree of randomness and the recovery error. Our proposed algorithm requires less total expected cost to achieve certain level of recovery accuracy. It also prolongs network lifetime due to its load balancing features. Extensive simulations on both synthetic and real data show that the proposed algorithm significantly outperforms benchmark algorithms.

Genetic algorithm-based optimisation of load-balanced routing for AMI with wireless mesh networks

The advanced metering infrastructure (AMI) in a smart grid contains hardware, software, and other electronic components connected through a communication infrastructure. AMI transfers meter-reading data between a group of smart meters and a utility centre. Herein, a wireless mesh network (WMN) with a random mesh topology is used to deploy the AMI communication network. In a WMN, paths are identified using a hybrid wireless mesh routing protocol (HWMP) with a load balancing feature called load aware-HWMP (LA-HWMP). These paths reduce the demand on links with a minimal air time metric; however, the delay in the data transmission of certain smart meters is high, given the large number of retransmissions caused by packet drop. To avert this problem and enhance the end-to-end delay, a genetic algorithm is applied on the LA-HWMP to obtain the optimal path. The optimisation process will result in the selection of paths with minimal delay. The genetic algorithm is developed with a rank-based selection, a two-point crossover, and a random reset mutation with a repair function to eliminate duplicate entries. The proposed method is compared with the HWMP, the LA-HWMP, and a state-of-the-art method that uses a combination of the ant colony algorithm and simulated annealing (ACA-SA) for AMI networks of different sizes. The obtained results show that the path identified by the proposed method yields a shorter delay and higher throughput than paths identified using the other methods.

Variants of the Low-Energy Adaptive Clustering Hierarchy Protocol: Survey, Issues and Challenges

A wireless sensor network (WSN) is a modern technology in radio communication. A WSN comprises a number of sensors that are randomly spread in a specific area for sensing and monitoring physical attributes that are difficult to monitor by humans, such as temperature, humidity, and pressure. Many problems, including data routing, power consumption, clustering, and selecting cluster heads (CHs), may occur due to the nature of WSNs. Various protocols have been conducted to resolve these issues. One of the important hierarchical protocols that are used to reduce power consumption in WSNs is low-energy adaptive clustering hierarchy (LEACH). This paper presents a comprehensive study of clustering protocols for WSNs that are relevant to LEACH. This paper is the first to emphasis on cluster formation and CHs selection methods and their strengths and weaknesses. A new taxonomy is presented to discuss LEACH variants on the basis of different classes, and the current survey is compared with other existing surveys. A complete comparison of the location, energy, complexity, reliability, multi–hop path, and load balancing characteristics of LEACH variants is conducted. Future research guidelines for CHs selection and cluster formation in WSNs are also discussed.

Load Balancing Metric Based Routing Protocol for Low Power and Lossy Networks (lbRPL)

IETF ROLL working Group standardized the IPv6 Routing protocol (RPL) for applications over low-power and lossy networks (LLNs). RPL constructs a Destination Oriented Direction Acyclic Graph (DODAG) to organize network topology. RPL shows fast network setup and good scalability. However, it may suffer from load imbalance due to diverse network traffic and heavy load on preferred or forwarding parents. To optimize the load balancing of routes in RPL, this paper proposes load balancing metric based routing protocol called lbRPL. We introduce a new routing metric for RPL called load balancing index (LBI), which exploits load balancing characteristics of RPL nodes to select more load balanced parents and routes. LBI includes ETX, Parent count (Pc) and Remaining Parent Energy (Pe) metrics to make routing decisions. Simulation results show that lbRPL improves network performance, stability and improved network life time to RPL.

A New Probabilistic Target Channel Selection Approach for Load Balancing in Cognitive Radio Networks

Cognitive radio is a technology targeting better utilization for the available frequency bands by allowing the unlicensed secondary users (SUs) to access the licensed frequency bands of the primary users (PUs). The SU vacates the channel at the presence of the PU and selects a target channel reactively or proactively to complete his interrupted transmission. Target channel sequence order can greatly affect the performance of the network in terms of delay, capacity, and utilization. The probabilistic sequence target channel selection approach was shown to distribute the load smoothly among the available channels resulting in good load balancing characteristics and increasing network capacity and stability for an appropriate distribution function. In this paper, a new proactive probabilistic sequence approach to determine the target channel for the interrupted SUs is proposed to achieve load balancing. The pre-emptive resume priority M/M/1 queueing network model is used to analyse the proposed approach to evaluate the latency and load balancing performance. Moreover, a new quantitative performance measure for balancing is introduced. The proposed approach is compared to other alternative spectrum handoff strategies and numerical results show the advantage of the proposed probabilistic sequence approach by improving balancing and channel capacity. Moreover, the extended data delivery time is improved at low network loads.

Implementing eco friendly highly reliable upload feature using multi 3G service

The current trend of eco friendly Internet access is preferred. In this research the understanding of eco friendly is minimum power consumption. The devices that are selected have operationally low power consumption and normally have no power consumption as they are hibernating during idle state. To have the reliability a router of a router that has internal load balancing feature will provide the improvement of previous research on multi 3G services for broadband lines. Previous studies emphasized on accessing and downloading information files from Public Cloud residing Web Servers. The demand is not only for speed but high reliability of access as well. High reliability will mean mitigating both direct and indirect high cost due to repeated attempts of uploading and downloading the large files. Nomadic and mobile computer users need viable solution. Following solution for downloading information has been proposed and tested. The solution is promising. The result is now extended to providing reliable access line by means of redundancy and automatic reconfiguration for uploading and downloading large information files to a Web Server in the Cloud. The technique is taking advantage of internal load balancing feature to provision a redundant line acting as a backup line. A router that has the ability to provide load balancing to several WAN lines is chosen. The WAN lines are constructed using multiple 3G lines. The router supports the accessing Internet with more than one 3G access line which increases the reliability and availability of the Internet access as the second line immediately takes over if the first line is disturbed.

Design of Network Analysis System Based on Stream Computing

The functions of the network analysis system include detection and analysis of network data stream. According to the results of the network analysis, we monitor the network accident and avoid the security risks. This can improve the network performance and increase the network availability. As the data flow in the network is constantly produced, the biggest characteristic of network analysis system is that it is a real-time system. Because of the high requirements of the network data analysis and network fault processing, the system requires very high processing efficiency of the real time data of network. Stream computing is a technique specifically for processing real-time data streams. Its idea is that the value of the data is reduced with the lapse of time, so as long as the data appearing, it must be processed as soon as possible. So we use the technology of stream computing to design network analysis system to meet the needs of real-time capability. Moreover, the stream computing framework has been widely welcomed in the field because of its good expansibility, ease of use and flexibility. In this paper, firstly, we introduce the characteristics of the data processing based on stream computing and the traditional data processing separately. We point out their difference and introduce the technique of stream computing. Then, we introduce the architecture of network analysis system designed base on the technique of stream computing. The architecture includes two main components that are logic processing layer and communication layer. We describe the characteristics of each component and functional characteristics in detail, and we introduce the system load balancing algorithm. Finally, by experiments, we verify the effectiveness of the system’s characteristics of dynamic expansion and load balancing.