Distributed gateway-based load balancing in software defined network

DSMLB: Dynamic switch-migration based load balancing for software-defined IoT network

Internet Download Manager (IDM) is one of the popular applications used by users to download various types of files from the internet. Load balancing is a technique used to increase the efficiency of using network resources by distributing data traffic evenly across several connection lines. This research aims to determine the extent to which IDM influences the effectiveness of load balancing as well as the influence of using Internet Download Manager (IDM) in the context of load balancing using the Policy-based Load Balancing (PCC), Equal Cost Multi-Path (ECMP), and Next-Hop Target methods. (NTH) on the MikroTik RB751 device. In this research, we used the PPDiOO method for systematic data collection and Quality of Services (QoS) analysis as a parameter for network data collection techniques. The research subject is the system or user who uses the device, while the research object is the impact of using IDM on load balancing performance. The main independent variable is the use of IDM, while the dependent variable is load balancing performance with different methods. Our findings show that using IDM consistently provides significant improvements in network performance compared to not using IDM.

The state-of-the-art scheduler of containerized cloud services considers load balance as the only criterion; many other important properties, including application performance, are overlooked. In the era of Big Data, however, applications evolve to be increasingly more data-intensive thus perform poorly when deployed on containerized cloud services. To that end, this paper aims to improve today's cloud service by taking application performance into account for the next-generation container schedulers. More specifically, in this work we build and analyze a new model that respects both load balance and application performance. Unlike prior studies, our model abstracts the dilemma between load balance and application performance into a unified optimization problem and then employs a statistical method to efficiently solve it. The most challenging part is that some sub-problems are extremely complex (for example, NP-hard), and heuristic algorithms have to be devised. Last but not least, we implement a system prototype of the proposed scheduling strategy for containerized cloud services. Experimental results show that our system can significantly boost application performance while preserving high load balance.

We highlight the advantages of dynamic wavelength allocation and path rerouting in hybrid optical flow-switched data center networks compared with static electronic packet-switched data center networks. Current electronic packet-switched networks perform routing based on lookup tables and shortest path routing over high speed optical connections. However, load balancing and path rerouting are not designed in coordination with the optical data plane. In this work, we design and analyze the performance of load balancing and resource allocation in hybrid flow data center architectures. In the hybrid flow architecture, the transport layer and control plane are responsible for load balancing, congestion control and reliable data delivery in wavelength division multiplexed networks. The transport layer abstracts the underlying network and data link layers from user applications. Applications are unaware of whether flows are transmitted via the electronic data plane, optical data plane, or both, where the relevant metric for applications is transaction delay. We show that the data center architecture that minimizes delay in the presence of elephant flows and dynamic loads is the hybrid flow architecture with dynamic resource allocation.

Routing and load balancing are two important factors that guarantee Quality of Service (QoS) in the Internet. Meanwhile Software Defined Networking (SDN) has introduced a new networking paradigm that make routing and load balancing policies implemented more quickly and flexibly in the system. In the SDN architecture, the "intelligence" of the network is centralized in the application layer and control layer instead of distributing in the network device in traditional network. Current studies using SDN-based routing and load balancing still stop at the scope of finding appropriate network resource for one single data flow but not considering network resource for all flows coming from different sources as the whole. In this paper, we propose a global load-balanced routing scheme, which can take advantages of global view of the SDN controller to make a global policy for routing and load balancing. The solution has been shown to outperform traditional algorithms and recent SDN-based artificial intelligence algorithms in terms of delay and network utilization.

In the area of cloud computing load balancing, the Particle Swarm Optimization (PSO) algorithm is neoteric and now praised highly, but recently a more neoteric algorithm which deploys the classifier into load balancing is presented. Besides, an algorithm called red-black tree which is aiming at improving the efficiency of resource dispatching is also praised. But the 3 algorithms all have different disadvantages which cannot be ignored. For example, the dispatch efficiency of PSO algorithm is not satisfying; although classifier and red-black tree algorithm improve the efficiency of dispatching tasks, the performance in load balancing is not that good, as a result the improved PSO algorithm is presented. Some researches are designed to get the advantages of new algorithm. First of all, the time complexity and performance for each algorithm in theory are computed; and then actual data which are generated in experiments are given to demonstrate the performance. And from the experiment result, it can be found that for the speed of algorithm itself PSO is the lowest, and the improved PSO solve this problem in some degree; improved PSO algorithm has the best performance in task solving and PSO is the second one, the red-black and Naive Bayes algorithm are much slower; PSO and improved PSO algorithm perform well in load balancing, while the other two algorithms do not do well.

Load balancing technique may improve the performance and scalability of the Internet greatly. Many researchers focus on intra-domain load balancing which distributes traffic over multiple paths or server farms in a single domain. However, there is little research on interdomain (such as inter-AS or inter-PoP) load balancing due to the loose relationship and complex policies among domains. It is important to study inter-domain load balancing under the trend toward a denser mesh of inter-domain connectivity within the Internet. In this paper, we propose a distributed and dynamic load balancing mechanism using a hashing-based algorithm to distribute traffic between domains which preserves per-flow packet ordering and supports unequal weighted distribution when dynamic adaptation is used. We also evaluate the performance of the mechanism.

This paper presents a migration mechanism for tasks running in nodes of a wireless sensor network (WSN). The WSN is consisted of digital signal processing nodes implemented with Blackfin microcomputers with Micrium μC OS III operating system. The migration mechanism allows load balancing, better scheduling of the tasks and allow cooperative signal processing at the whole network and it is useful in case of the node failures as well. This mechanism uses the Micrium operating system primitives in order to decide when the task migration will be started and to control the whole migration process. The communication network is provided by Lightweight Internet Protocol (LwIP) protocol. A performance evaluation (load balance, migration time and missed deadlines ratio) is performed. The paper demonstrates viable implementation of tasks migration in embedded systems and illustrates a better performance of the load balancing and the tasks scheduling.

Lightweight container technology has emerged as a fundamental component of cloud-native computing, with the deployment of containers and the balancing of loads on virtual machines representing significant challenges. This paper presents an optimization strategy for container deployment that consists of two stages: coarse-grained and fine-grained load balancing. In the initial stage, a greedy algorithm is employed for coarse-grained deployment, facilitating the distribution of container services across virtual machines in a balanced manner based on resource requests. The subsequent stage utilizes a genetic algorithm for fine-grained resource allocation, ensuring an equitable distribution of resources to each container service on a single virtual machine. This two-stage optimization enhances load balancing and resource utilization throughout the system. Empirical results indicate that this approach is more efficient and adaptable in comparison to the Grey Wolf Optimization (GWO) Algorithm, the Simulated Annealing (SA) Algorithm, and the GWO-SA Algorithm, significantly improving both resource utilization and load balancing performance on virtual machines.

Software-defined networking (SDN) is a state-of-the-art architectural approach to network management that allows more flexible management of a complicated large-scale network. For providing services with high performance, availability, and scalability, server clustering is widely used. This study develops a server cluster based on OpenFlow, which is the most common SDN protocol. To utilize cluster resources effectively, this study focuses on the server-side load-balancing issue and increases the performance for a cluster of servers with different processing capabilities. We have proposed a novel load balancing algorithm, named dynamic weighted random selection (DWRS). DWRS considers the real-time server loads when dispatching requests to servers. Underutilized servers are assigned higher weights, so they have a higher possibility of being selected to process requests. To increase system performance, we modify the flow of packet handling in the Floodlight controller. A multi-threaded technique is used to effectively utilize the parallel processing capability of multiple cores, which speeds up the processing of packet-in messages when selecting the target server. This prevents the controller from becoming overloaded, especially when a large number of packet-in messages arrive. To demonstrate effectiveness and feasibility, we deploy our experimental environment using real hardware, instead of using Mininet to set up a virtual SDN testbed. Experimental results show that the proposed DWRS with multi-threaded implementation outperforms other policies, especially in a heterogeneous environment.

The increasing complexity of web applications and the growing number of concurrent users require load-balancing mechanisms that can maintain system performance and service reliability. This study aims to evaluate and compare the performance of NGINX and Caddy as load balancers for WordPress applications deployed in a Podman-based containerized environment. A quantitative experimental approach is employed by measuring key Quality of Service (QoS) metrics, including request rate, response time, and error rate, across multiple load scenarios and load-balancing algorithms, namely round-robin and least connection. The experimental results indicate that under low to medium load conditions, both NGINX and Caddy exhibit comparable performance. However, under high-load conditions, performance differences become significant. Caddy, particularly when using the least-connection algorithm, can maintain substantially higher throughput, achieving an aggregate request rate of approximately 2349.55 req/s, compared to NGINX, which records approximately 426.02 req/s under comparable configurations. In terms of response time, the P90 response time for Caddy is approximately 289.79 ms, whereas NGINX exceeds 4 seconds. Furthermore, Caddy demonstrates superior reliability with an error rate of 0% across all testing scenarios, while NGINX experiences extremely high request failure rates under extreme load conditions. These findings suggest that Caddy, particularly when combined with the least-connection algorithm, provides a more adaptive and stable load-balancing solution for dynamic, containerized web applications with fluctuating traffic patterns. Nevertheless, this study is conducted in a controlled environment using baseline configurations without advanced optimizations; therefore, the results represent a relative performance comparison rather than an absolute benchmark for all production environments. Future research is recommended to investigate advanced configurations, different web application types, and more diverse workload scenarios to achieve a more comprehensive understanding of load balancer performance in containerized environments.

Finite difference methods are commonplace in scientific computing. Despite their apparent regularity, they often exhibit load imbalance that damages their efficiency. We characterize the spatial and temporal load imbalance of Ondes3D, a seismic wave propagation simulator. We reveal that this imbalance originates from the nature of the input data and from low-level CPU optimizations. Such dynamic imbalance should therefore be quite common and is intractable by any static approach or classical code reorganization. An effective solution, with few code modifications, combines domain over-decomposition and dynamic load balancing (e.g., with AMPI), migrating data and computation at the granularity of an MPI rank. It generally requires a careful tuning of the over-decomposition level, the load balancing heuristic and frequency. These choices are quite dependent on application and platform characteristics. In this paper, we propose a methodology that leverages the capabilities of the SimGrid framework to conduct such study at low experimental cost. It combines emulation, simulation, and application modeling that requires minimal code modification and yet manages to capture both spatial and temporal load imbalance, faithfully predicting its overall performance. We compare simulation and real executions results and show how our strategy can be used to determine the best load balancing configuration for a given application/hardware configuration.

With the rapid development of information technology, enormous volumes of data is being generated by many enterprises at all times. A reasonable storage of these large scale data to reduce cost and achieve internal data sharing and collaboration has always been a challenge for enterprises. Cloud storage technology, as an important branch in the field of cloud computing, is becoming a trend to solve this problem. At present, there have been some works on cloud storage, but most of them focus on individual users. Even if there are a few enterprise level products, they still cannot meet the actual needs of enterprises. To address these problems, in this paper, we propose a flexible enterprise-oriented cloud storage system (ECStor) based on GlusterFS. ECStor provides a directory-level fine-grained access control solution, so that storage space can be allocated flexibly according to different role of users or IP addresses. Meanwhile, we achieved a load balancer to distribute user load to NFS servers on different storage nodes, and also implemented a friendly management system interface based on B/S architecture. Finally, we provide an easy-to-use client-side software on Windows platform. We deployed ECStor to the data center, and evaluated the performance of LIST and load balance and found that the latency of LIST acted out no significant losses compared with the original GlusterFS, and achieved a good load balance.

The advance of the GPU hardware architecture has made GPUs attractive devices for general- purpose computing. Modern GPUs are equipped with an increasing number of cores, a flexible memory hierarchy, and a large memory capacity. While the computational power of modern GPU devices has allowed their introduction in high-performance computing (HPC) clusters and the efficient processing of ever larger workloads, existing software components for HPC clusters still offer basic support for hardware heterogeneity and often cause performance limitations in the presence of GPU devices. In particular, two kinds of limitations are associated with these software components: runtime support and programmability. We found that these limitations are due to the fact that existing software frameworks for heterogeneous clusters treat GPUs as dedicated coprocessor devices. In this dissertation, we propose two software frameworks for addressing the performance and hardware underutilization issues found in heterogeneous CPU-GPU clusters as well as increasing their programmability. Our frameworks provide a uniform view of compute resources and treat CPUs and GPUs equally as first-class resources, allowing efficient management of heterogeneous compute resources. First, we propose a hierarchical scheduling framework consisting of a node-level runtime and a cluster-level scheduler that provides abstraction of heterogeneous compute resources at different granularities. This hierarchical framework targets existing applications and does not require their modification. In the node-level runtime, we identify and design mechanisms, such as virtual GPUs, GPU virtual memory, dynamic load balancing and pre-emption, which are necessary to support efficient sharing and load balancing schemes for GPUs within a compute node. In the cluster-level scheduler, we introduce mechanisms to abstract compute nodes and perform load balancing in concert with the node-level runtime. Our hierarchical scheduling framework allows supporting different load balancing policies and does not require additional inputs (such as profiling information) from users. Second, we propose a programming framework based on a novel memory and execution model. Our memory model hides disjoint addressing spaces (corresponding to different CPUs, GPUs and compute nodes) and provides a view of a single virtual memory space that can be accessed by all compute resources in a heterogeneous cluster. Our execution model provides uniform access to compute resources and allows our framework to treat all CPUs and GPUs equally and to access data in the virtual memory space.

In 5G scenarios, the dynamic resource allocation of network slicing is crucial for quality-of-service (QoS) guaranteed under fluctuating traffic demands in rapidly changing communication environments. In this paper, we propose a novel QoS guaranteed joint resource allocation framework for NR with supplementary uplink (SUL) called QGJRA-SUL, where three parameters of SUL admission, TDD pattern, and band slicing scheme are jointly optimized. The framework is driven by a well-designed deep reinforcement learning agent. By combining the activation functions tanh and softmax, the agent can jointly optimize three parameters at the same time. Under the original problem of QoS satisfaction rate maximization, we introduce the load unbalance degree of slices into the reward function as a penalty term. The simulation results show that the framework can guarantee the QoS satisfaction rate well and balance the load of slices. QGJRA-SUL can accommodate 15% more user equipments (UEs) with the same QoS satisfaction rate than that of a traditional single-band solution without SUL, and achieve a 73% increase in the performance of load balancing than that without a load balancing mechanism near the full load.