Computing Load Balancing Research Articles

Cloud Computing Load Balancing Algorithms' Difficulties

Asian Journal of Organic & Medicinal Chemistry includes communications, full papers (articles), short papers, green chemistry, reviews, and perspectives. Specifically, these include synthesis, synthetic methods and regents, mechanism of reactions, biologically active compounds with their possible mechanism of action, natural products, in vitro and in vivo tests of compounds, computational chemistry, physical-analytical-inorganic chemistry, nanotechnology, drug carriers, proteomics and genomics. Rapid synthesis of new compounds following eco-friendly methods and their functions as medicinally active agents is certainly the central part of chemical and medicinal research in modern times. On the basis of this fact, significant emphasis is placed to rapid publication and disseminates original discoveries in this new journal.

Reduce Operations: Send Volume Balancing While Minimizing Latency

Communication hypergraph model was proposed in a two-phase setting for encapsulating multiple communication cost metrics (bandwidth and latency), which are proven to be important in parallelizing irregular applications. In the first phase, computational-task-to-processor assignment is performed with the objective of minimizing total volume while maintaining computational load balance. In the second phase, communication-task-to-processor assignment is performed with the objective of minimizing total number of messages while maintaining communication-volume balance. The reduce-communication hypergraph model suffers from failing to correctly encapsulate send-volume balancing. We propose a novel vertex weighting scheme that enables part weights to correctly encode send-volume loads of processors for send-volume balancing. The model also suffers from increasing the total communication volume during partitioning. To decrease this increase, we propose a method that utilizes the recursive bipartitioning framework and refines each bipartition by vertex swaps. For performance evaluation, we consider column-parallel SpMV, which is one of the most widely known applications in which the reduce-task assignment problem arises. Extensive experiments on 313 matrices show that, compared to the existing model, the proposed models achieve considerable improvements in all communication cost metrics. These improvements lead to an average decrease of 30 percent in parallel SpMV time on 512 processors for 70 matrices with high irregularity.

Minimizing Deadline Misses and Total Run-time with Load Balancing for a Connected Car Systems in Fog Computing

Cloud computing helps in providing the applications with a few number of resources that are used to unload the tasks. But there are certain applications like coordinated lane change assistance which are helpful in cars that connects to internet has strict time constraints, and it may not be possible to get the job done just by unloading the tasks to the cloud. Fog computing helps in reducing the latency i.e the computation is now done in local fog servers instead of remote datacentres and these fog servers are connected to the nearby distance to clients. To achieve better timing performance in fog computing load balancing in these fog servers is to be performed in an efficient manner.The challenges in the proposed application includes the number of tasks are high, client mobility and heterogeneous nature of fog servers. We use mobility patterns of connected cars and load balancing is done periodically among fog servers. The task model presented here in this paper solves scheduling problem and this is done at the server level and not on the device level. And at last, we present an optimization problem formulation for balancing the load and for reducing the misses in deadline, also the time required for running the task in these cars will be minimized with the help of fog computing. It also performs better than somecommon algorithms such as active monitoring, weighted round robin and throttled load balancer.

Parallelization of Implementations of Purely Sequential Algorithms

The work is dedicated to the parallelization of programs in especially difficult cases when the used algorithm is purely sequential, there are no parallel alternatives to this algorithm, and its execution time is unacceptably high. Various parallelization methods for software implementations of such algorithms and resulting computational load balancing are considered that make it possible to considerably accelerate the execution of application programs using purely sequential algorithms. The proposed methods are illustrated with examples of their application to two algorithms used in a dynamic binary code analysis toolset. The main goal of this paper is to show that the use of a purely sequential algorithm in a software implementation does not necessarily imply that its execution is inevitably sequential. The proposed methods of parallelizing implementations of such algorithms and balancing the resulting computational load can help develop efficient parallel programs that fully utilize the hardware capabilities of modern computers.

A Fog Computing-based IoT Framework for Precision Agriculture

The challenge of analyzing and processing a huge amount of data is becoming increasingly important in this fourth industrial revolution era. In this scenario, Cloud Computing and Internet of Things (IoT) allow to build up an interconnected network of smart things. These two paradigms do not allow solving the Computing problems yet. Fog Computing aims at moving the processing abilities closer to the end users, avoiding an excessive exploitation of Cloud resources, further reducing computational loads. In this work, we propose a Fog-based IoT framework, which exploits the two-tier Fog and their resources, reducing the transmitted data to the Cloud, improving the computational load balancing and reducing the waiting times. The proposed Fog Computing approach is applied to the emerging area of precision agriculture, including all the techniques of agricultural land management. Furthermore, based on this framework, we have simulated and highlighted how the two-tier Fog Computing approach is able to reduce significantly the amount of transmitted data to the Cloud. We also propose and describe an application prototype, based on the previous framework, able to manage and monitor farmland, with a strong impact on both the business and environmental performance.

Computational Load Balancing on the Edge in Absence of Cloud and Fog

Mobile Cloud Computing or Fog computing refers to offloading computationally intensive algorithms from a mobile device to the cloud or an intermediate cloud in order to save resources e.g., time and energy in the mobile device. This paper proposes new solutions for situations when the cloud or fog is not available. First, the sensor network is modelled using a network of queues, then a linear programming technique is used to make scheduling decisions. Various centralized and distributed algorithms are then proposed, which improves overall system performance. Extensive simulations show slightly higher energy usage in comparison to the baseline non-offloading case, however, the job completion rate is significantly improved, the efficiency score metric shows the extra energy usage is justified. The algorithms have been simulated in various environments including high and low bandwidth, partial connectivity, and different rate of information exchanges to study the pros and cons of the proposed algorithms.

Reverse Time Migration of Elastic Waves Using the Pseudospectral Time-Domain Method

Reverse time migration (RTM), especially that for elastic waves, consumes massive computation resources which limit its wide applications in industry. We suggest to use the pseudospectral time-domain (PSTD) method in elastic wave RTM. RTM using PSTD can significantly reduce the computational requirements compared with RTM using the traditional finite difference time domain method (FDTD). In addition to the advantage of low sampling rate with high accuracy, the PSTD method also eliminates the periodicity (or wraparound) limitation caused by fast Fourier transform in the conventional pseudospectral method. To achieve accurate results, the PSTD method needs only about half the spatial sampling rate of the twelfth-order FDTD method. Thus, the PSTD method can save up to 87.5% storage memory and 90% computation time over the twelfth-order FDTD method. We implement RTM using PSTD for elastic wave equations and accelerate it by Open Multi-Processing technology. To keep the computational load balance in parallel computation, we design a new PML layout which merges the PML in both ends of an axis together. The efficiency and imaging quality of the proposed RTM is verified by imaging on 2D and 3D models.

Heterogeneous Ultradense Networks with NOMA: System Architecture, Coordination Framework, and Performance Evaluation

Heterogeneous ultradense networks (H-UDNs) are one key enabler for fifth-generation (5G) wireless networks and beyond to satisfy the explosive growth of mobile data traffic, which exploits spatial reuse of scarce spectrum by deploying massive base stations (BSs) to boost network capacity and enhance network coverage. In this article, we present the system architecture for 5G H-UDNs, consisting of virtualized integrated ground-air-space radio access networks (RANs) and core networks and study network coordination for virtualized H-UDN to efficiently manage computing resources and intercell interference. We look at a cloud-fog-computing coordination framework for efficient computing resource management by achieving reasonable computing task distribution and transfer; computing load balance for computing tasks among virtual computing resources to improve network performance and computing resource efficiency; and a macro-small cell coordination framework for virtualized H-UDN with nonorthogonal multiple access (NOMA) to efficiently manage intercell interference and improve network performance. The joint macro-small enhanced intercell interference coordination (eICIC) and small-small coordinated multipoint joint transmission (CoMP-JT) scheme can efficiently eliminate macro-small intercell interference and utilize small-small intercell interference.

Распараллеливание реализаций сугубо последовательных алгоритмов

The work is dedicated to the topic of parallelizing programs in especially difficult cases - when the used algorithm is purely sequential, there are no parallel alternatives to the algorithm used, and its execution time is unacceptably high. Various parallelization methods for software implementations of such algorithms and resulting computational load balancing are considered, allowing to obtain significant performance acceleration for application programs using purely sequential algorithms. The above methods are illustrated by the practice of their application to two algorithms used in a dynamic binary code analysis toolset. The main goal of this paper is to show that the use of a purely sequential algorithm in a software implementation does not necessarily imply inevitability of its sequential execution. The proposed methods of parallelizing implementations of such algorithms and balancing the resulting computational load can help to develop efficient parallel program that fully utilize the hardware capabilities of modern computing systems.

Towards Long-View Computing Load Balancing in Cluster Storage Systems

In large-scale computing clusters, when the server storing a task's requested data does not have sufficient computing capacity for the task, current job schedulers either schedule the task to the closest server and transmit to it the requested data, or let the task wait until the server has sufficient computing capacity. The former solution generates network load while the latter solution increases task delay. To handle this problem, load balancing methods are needed to reduce the number of overloaded servers due to computing workloads. However, current load balancing methods do not aim to balance the computing load for the long term. Through trace analysis, we demonstrate the diversity of computing workloads of different tasks and the necessity of balancing the computing workloads among servers. Then, we propose a cost-efficient Computing load Aware and Long-View load balancing approach ( CALV ). CALV is novel in that it achieves long-term computing load balance by migrating out an overloaded server data blocks contributing more computing workloads when the server is more overloaded and contribute less computing workloads when the server is more underloaded at different epochs during a time period. Based upon the task schedules, we further propose a task reassignment algorithm that reassigns tasks from an overloaded server to other data servers of the tasks to make it non-overloaded before CALV is conducted. The above methods are for the tasks whose submission times and execution latencies can be predicted. To handle unexpected tasks or insufficiently accurate predictions, we propose a dynamic load balancing method, in which an overloaded server dynamically redirects tasks to other data servers of the tasks, or replicates the tasks’ requested data to other servers and redirects the tasks to those servers in order to become non-overloaded. Finally, we propose a proximity-aware tree based distributed load balancing method to reduce the reallocation cost and improve the scalability of CALV . Trace-driven experiments in simulation and a real computing cluster show that CALV outperforms other methods in terms of balancing the computing workloads and cost efficiency.

Reverse Time Migration of Elastic Waves Using the Pseudospectral Time-Domain Method

Reverse time migration (RTM), especially that for elastic waves, consumes massive computation resources which limit its wide applications in industry. We suggest to use the pseudospectral time-domain (PSTD) method in elastic wave RTM. RTM using PSTD can significantly reduce the computational requirements compared with RTM using the traditional finite difference time domain method (FDTD). In addition to the advantage of low sampling rate with high accuracy, the PSTD method also eliminates the periodicity (or wraparround) limitation caused by fast Fourier transform in the conventional pseudospectral method. To achieve accurate results, the PSTD method needs only about half the spatial sampling rate of the twelfth-order FDTD method. Thus, the PSTD method can save up to 87.5% storage memory and 90% computation time over the twelfth-order FDTD method. We implement RTM using PSTD for elastic wave equations and accelerate it by Open Multi-Processing technology. To keep the computational load balance in parallel ...

Scheduling based on a dynamic resource connection

The practical using of distributed computing systems associated with many problems, including troubles with the organization of an effective interaction between the agents located at the nodes of the system, with the specific configuration of each node of the system to perform a certain task, with the effective distribution of the available information and computational resources of the system, with the control of multithreading which implements the logic of solving research problems and so on. The article describes the method of computing load balancing in distributed automatic systems, focused on the multi-agency and multi-threaded data processing. The scheme of the control of processing requests from the terminal devices, providing the effective dynamic scaling of computing power under peak load is offered. The results of the model experiments research of the developed load scheduling algorithm are set out. These results show the effectiveness of the algorithm even with a significant expansion in the number of connected nodes and zoom in the architecture distributed computing system.

LBMIC: communication-aware load balancing in distributed ASMs with evolving social networks

Multi-agent-based simulation for artificial stock market (ASM) is an important method in behavioural finance. The social network in ASM will influence the coordination and decision making of the intelligent agents. To improve the performance of an ASM with evolving social networks in a distributed computing environment, the computational load balancing and inter-nodes communication should be considered jointly. This paper proposes a scheduling algorithm called LBMIC to partition the agents onto different computing nodes while keeping the degree of load imbalance lower than a given threshold with minimized inter-nodes communication between agents. LBMIC models the scheduling into a graph partitioning problem and uses the multi-level graph partitioning algorithm to achieve an efficient scheduling. When the network evolves, LBMIC refines the partitioning by migrating parts of the agents. The experiments indicate that LBMIC can efficiently improve the performance of communication-intensive ASMs by both initial partitioning and refining partitioning.

Implementation of the Open Source Virtualization Technologies in Cloud Computing

The “Virtualization and Cloud Computing” is a recent buzzword in the digital world. Behind this fancy poetic phrase there lies a true picture of future computing for both in technical and social perspective. Though the “Virtualization and Cloud Computing are recent but the idea of centralizing computation and storage in distributed data centres maintained by any third party companies is not new but it came in way back in 1990s along with distributed computing approaches like grid computing, Clustering and Network load Balancing. Cloud computing provide IT as a service to the users on-demand basis. This service has greater flexibility, availability, reliability and scalability with utility computing model. This new concept of computing has an immense potential in it to be used in the field of e-governance and in the overall IT development perspective in developing countries like Bangladesh.

Research on cloud computing load balancing based on information entropy

This article is designed to solve the problem of load balancing in cloud computing. Virtual machine migration is a way of resolving load balancing. This study firstly analyzes load selecting, load gathering, source virtual machine selecting and ect in detail. The most important research is the target machine selecting and a selecting algorithm is proposed based on information entropy. Experimental results reveal that the algorithm can not only achieve load balancing but also improve the performance of the system.

Computational Load Balancing Algorithm for Parallel Knapsack Packing Tree Traversal

The paper considers efficient computational load distribution for the exact parallel algorithm for the knapsack problem based on packing tree search. We propose an algorithm that provides for static and dynamic computational load balancing for the problem in question.

On the use of Schwarz–Christoffel conformal mappings to the grid generation for global ocean models

Abstract. In this article we propose two grid generation methods for global ocean general circulation models. Contrary to conventional dipolar or tripolar grids, the proposed methods are based on Schwarz–Christoffel conformal mappings that map areas with user-prescribed, irregular boundaries to those with regular boundaries (i.e., disks, slits, etc.). The first method aims at improving existing dipolar grids. Compared with existing grids, the sample grid achieves a better trade-off between the enlargement of the latitudinal–longitudinal portion and the overall smooth grid cell size transition. The second method addresses more modern and advanced grid design requirements arising from high-resolution and multi-scale ocean modeling. The generated grids could potentially achieve the alignment of grid lines to the large-scale coastlines, enhanced spatial resolution in coastal regions, and easier computational load balance. Since the grids are orthogonal curvilinear, they can be easily utilized by the majority of ocean general circulation models that are based on finite difference and require grid orthogonality. The proposed grid generation algorithms can also be applied to the grid generation for regional ocean modeling where complex land–sea distribution is present.

Towards a multi-QoS human-centric cloud computing load balance resource allocation method

In the large-scale clustering resource pool as human-centric cloud computing, peer load balance not only improves overall system efficiency, but also saves energy. As various factors should be considered in resource scheduling and each has different emphasis, resource allocation method adapted by different scene also has respective criteria. Based on resource allocation techniques, the multi-QoS load balance resource allocation method (MQLB-RAM) was proposed in the paper. It combines needs of users and service providers to constitute multi-QoS indexes. The needs from cost, system and network were met by quantitative analysis on load balancing using real-time load of peers. The algorithm also compares weight of each index in peer to match need and resource, so as to achieve the target of ensuring load balance, making full use of resources and saving money. Simulation experiment with CloudSim shows that the MQLB-RAM can achieve balance among load, resource access performance and cost.

Software Pipeline–Based Partitioning Method with Trade-Off between Workload Balance and Communication Optimization

For a multiprocessor System-on-Chip (MPSoC) to achieve high performance via parallelism, we must consider how to partition a given application into different components and map the components onto multiple processors. In this paper, we propose a software pipeline–based partitioning method with cyclic dependent task management and communication optimization. During task partitioning, simultaneously considering computation load balance and communication optimization can cause interference, which leads to performance loss. To address this issue, we formulate their constraints and apply an integer linear programming approach to find an optimal partitioning result — one that requires a trade-off between these two factors. Experimental results on a reconfigurable MPSoC platform demonstrate the effectiveness of the proposed method, with 20% to 40% performance improvements compared to a traditional software pipeline–based partitioning method.

A parallel implementation of an off-lattice individual-based model of multicellular populations

As computational models of multicellular populations include ever more detailed descriptions of biophysical and biochemical processes, the computational cost of simulating such models limits their ability to generate novel scientific hypotheses and testable predictions. While developments in microchip technology continue to increase the power of individual processors, parallel computing offers an immediate increase in available processing power. To make full use of parallel computing technology, it is necessary to develop specialised algorithms. To this end, we present a parallel algorithm for a class of off-lattice individual-based models of multicellular populations. The algorithm divides the spatial domain between computing processes and comprises communication routines that ensure the model is correctly simulated on multiple processors. The parallel algorithm is shown to accurately reproduce the results of a deterministic simulation performed using a pre-existing serial implementation. We test the scaling of computation time, memory use and load balancing as more processes are used to simulate a cell population of fixed size. We find approximate linear scaling of both speed-up and memory consumption on up to 32 processor cores. Dynamic load balancing is shown to provide speed-up for non-regular spatial distributions of cells in the case of a growing population.