Novel dynamic load balancing algorithm for cloud-based big data analytics

Big data analytics in cloud environments introduces challenges such as real-time load balancing besides security, privacy, and energy efficiency. This paper proposes a novel load balancing algorithm in cloud environments that performs resource allocation and task scheduling efficiently. The proposed load balancer reduces the execution response time in big data applications performed on clouds. Scheduling, in general, is an NP-hard problem. Our proposed algorithm provides solutions to reduce the search area that leads to reduced complexity of the load balancing. We recommend two mathematical optimization models to perform dynamic resource allocation to virtual machines and task scheduling. The provided solution is based on the hill-climbing algorithm to minimize response time. We evaluate the performance of proposed algorithms in terms of response time, turnaround time, throughput metrics, and request distribution with some of the existing algorithms that show significant improvements.

     

Adaptive scheduling for parallel tasks with QoS satisfaction for hybrid cloud environments

A hybrid cloud integrates private clouds and public clouds into one unified environment. For the economy and the efficiency reasons, the hybrid cloud environment should be able to automatically maximize the utilization rate of the private cloud and minimize the cost of the public cloud when users submit their computing jobs to the environment. In this paper, we propose the Adaptive-Scheduling-with-QoS-Satisfaction algorithm, namely AsQ, for the hybrid cloud environment to raise the resource utilization rate of the private cloud and to diminish task response time as much as possible. We exploit runtime estimation and several fast scheduling strategies for near-optimal resource allocation, which results in high resource utilization rate and low execution time in the private cloud. Moreover, the near-optimal allocation in the private cloud can reduce the amount of tasks that need to be executed on the public cloud to satisfy their deadline. For the tasks that have to be dispatched to the public cloud, we choose the minimal cost strategy to reduce the cost of using public clouds based on the characteristics of tasks such as workload size and data size. Therefore, the AsQ can achieve a total optimization regarding cost and deadline constraints. Many experiments have been conducted to evaluate the performance of the proposed AsQ. The results show that the performance of the proposed AsQ is superior to recent similar algorithms in terms of task waiting time, task execution time and task finish time. The results also show that the proposed algorithm achieves a better QoS satisfaction rate than other similar studies.     

A delay-based dynamic scheduling algorithm for bag-of-task workflows with stochastic task execution times in clouds

Bag-of-Tasks (BoT) workflows are widespread in many big data analysis fields. However, there are very few cloud resource provisioning and scheduling algorithms tailored for BoT workflows. Furthermore, existing algorithms fail to consider the stochastic task execution times of BoT workflows which leads to deadline violations and increased resource renting costs. In this paper, we propose a dynamic cloud resource provisioning and scheduling algorithm which aims to fulfill the workflow deadline by using the sum of task execution time expectation and standard deviation to estimate real task execution times. A bag-based delay scheduling strategy and a single-type based virtual machine interval renting method are presented to decrease the resource renting cost. The proposed algorithm is evaluated using a cloud simulator ElasticSim which is extended from CloudSim. The results show that the dynamic algorithm decreases the resource renting cost while guaranteeing the workflow deadline compared to the existing algorithms.     

Online cost-rejection rate scheduling for resource requests in hybrid clouds

The hybrid cloud computing model has attracted considerable attention in recent years. Due to security and controllability of private cloud, some resource requests are required to be scheduled in private parts of hybrid cloud. However, these requests could be rejected because of the resource limitation of private parts of hybrid cloud. In this paper, all resource requests are classified into two categories, i.e., the special requests required to process in private cloud, and the normal requests insensitive in private or public clouds. By considering the normal requests undivided to dispatch a cloud and divided to dispatch different clouds, we propose the online cost-rejection rate scheduling strategy (OCS) for the normal requests undivided, and the online scheduling strategy for some normal requests divided (OCS-divided), which could make suitable requests placement decisions in real-time and minimize the cost of renting public cloud resources with a low rate of rejected requests. Then, we transform both online models into two one-shot optimization problems by taking advantage of Lyapunov optimization techniques, and employ the optimal decay algorithm to solve the one-shot problems. The simulation results demonstrate that our scheduling strategies can achieve the trade-off between cost and rejection rate, and process the real-time resource requests in hybrid clouds. OCS-divided can achieve an average cost saving of about 25% compared with OCS and maximize the resource utilization.     

混合云环境中数据敏感工作流调度

混合云环境下调度包含敏感数据的工作流主要考虑在满足数据安全性以及工作流截止时间的前提下,对工作流任务在混合云上进行分配,实现计算资源与任务的映射,并优化调度费用。采用了整数规划来建模求解包含数据敏感性、截止时间和调度费用3种约束条件的混合云工作流调度问题,同时为优化模型求解速度,基于“帕雷托最优”原理对工作流任务在混合云上的分配方案进行筛选以减小模型求解规模。实验表明,优先排除不合理的任务分配方案可有效减小整数规划模型的求解规模,缩短模型计算时间,在产生较小误差的情况下获得较优的调度结果。     

安全服务链中虚拟网络功能分配与调度算法研究

安全服务链中的虚拟网络功能（virtual network function,VNF）将传统网络安全功能与硬件设备解耦,使得服务功能的部署更具动态性和可扩展性。然而,VNF向节点的合理分配以及节点上VNF的高效调度问题仍亟待解决。为此,基于软件定义网络（software defined network,SDN）和网络功能虚拟化（network function virtualization,NFV）环境,提出基于优化算法的解决方案。首先,对资源分配与调度问题进行举例并形式化定义问题的优化目标;其次,提出基于贪心算法的资源分配方案和基于混合蜂群算法的资源调度方案,统一协调解决VNF的资源分配与调度问题。最后,设计仿真实验,验证所提算法的时间复杂性和在总资源成本和总服务收益方面的提升;同时,对比混合蜂群算法和传统蜂群算法,结果显示前者具有更快的收敛速度。     

基于多种云环境的任务调度算法综述

任务调度算法是云计算资源分配部署的核心方法。针对当前云计算发展面临的任务需求和数据量指数级增长的问题,重点对任务调度算法进行了系统的梳理和归纳,以云环境为分类依据,研究分析了单云、联盟云、混合云、多云四类调度算法。在单云环境中,从传统启发式、元启发式以及混合式任务调度算法角度进行阐述。在联盟云、混合云、多云环境中,从工作流和独立任务调度算法角度进行阐述。通过比较,总结了现有算法的优点、缺点以及优化性能,并形成结论性意见和开放性问题,为未来对容器云、数据云以及兼顾资源分配与任务调度算法的研究奠定基础。     

Scheduling jobs with deterioration effect and controllable processing time

This article considered the single machine scheduling with controllable processing time (resource allocation) and deterioration effect concurrently. It discussed the minimization of three objectives, which involve the weighted sum of the makespan and the total resource consumption cost, the total resource consumption cost under the condition that the makespan (total flow time) is restricted to a fixed constant and the optimal resource allocation and the optimal job sequence is what we need to make decision. Considering the makespan constraint, it proved that these problems can be solved in polynomial time. A special case of the last problem can be solved in polynomial time with respect to the total flow time constraint.

     

Cooperative game-based distributed resource allocation in horizontal dynamic cloud federation platform

Distributed resource allocation is a very important and complex problem in emerging horizontal dynamic cloud federation (HDCF) platforms, where different cloud providers (CPs) collaborate dynamically to gain economies of scale and enlargements of their virtual machine (VM) infrastructure capabilities in order to meet consumer requirements. HDCF platforms differ from the existing vertical supply chain federation (VSCF) models in terms of establishing federation and dynamic pricing. There is a need to develop algorithms that can capture this complexity and easily solve distributed VM resource allocation problem in a HDCF platform. In this paper, we propose a cooperative game-theoretic solution that is mutually beneficial to the CPs. It is shown that in non-cooperative environment, the optimal aggregated benefit received by the CPs is not guaranteed. We study two utility maximizing cooperative resource allocation games in a HDCF environment. We use price-based resource allocation strategy and present both centralized and distributed algorithms to find optimal solutions to these games. Various simulations were carried out to verify the proposed algorithms. The simulation results demonstrate that the algorithms are effective, showing robust performance for resource allocation and requiring minimal computation time.     

Single-machine scheduling with learning effect and resource-dependent processing times

We consider resource allocation scheduling with learning effect in which the processing time of a job is a function of its position in a sequence and its resource allocation. The objective is to find the optimal sequence of jobs and the optimal resource allocation separately. We concentrate on two goals separately, namely, minimizing a cost function containing makespan, total completion time, total absolute differences in completion times and total resource cost; minimizing a cost function containing makespan, total waiting time, total absolute differences in waiting times and total resource cost. We analyse the problem with two different processing time functions. For each combination of these, we provide a polynomial time algorithm to find the optimal job sequence and resource allocation.     

Integer linear programming-based multi-objective scheduling for scientific workflows in multi-cloud environments

Scientific communities are motivated to schedule the data-intensive scientific workflows in multi-cloud environments, where considerable diverse resources are provided by multiple clouds and resource limitation imposed by individual clouds is overcome. However, this scheduling involves two conflicting objectives: minimizing cost and makespan. In general, dealing with such conflicting criteria is a difficult task. But fortunately recent efficient methods for solving multi-objective optimization problems motivated us to provide a multi-objective model considering minimization of cost and makespan as objectives. For solving this model, we use different scalarization procedures such as weighted-sum, Benson's scalarization and weighted min–max under different scenarios. Moreover, we investigate the stability of obtained solutions and propose a new approach for determining the most stable solution related to weighted-sum and weighted min–max as post-optimality analysis. Results indicate that our proposed weighted-sum approach outperforms the previously developed methods in terms of hypervolume.

     

Resource-dependent scheduling with deteriorating jobs and learning effects on unrelated parallel machine

The focus of this paper is to analyze unrelated parallel-machine resource allocation scheduling problem with learning effect and deteriorating jobs. The goal is to find the optimal sequence of jobs and the optimal resource allocation separately for minimizing the cost function including the total load, the total completion time, the total absolute deviation of completion time and the total resource cost. We show that the problem is polynomial time solvable if the number of machines is a given constant.     

An energy-efficient scheduling scheme for time-constrained tasks in local mobile clouds

Mobile Cloud Computing (MCC) enables mobile devices to use resource providers other than mobile devices themselves to host the execution of mobile applications. Various mobile cloud architectures and scheduling algorithms have been studied recently. However, how to utilize MCC to enable mobile devices to run complex real-time applications while keeping high energy efficiency remains a challenge. In this paper, firstly, we introduce the local mobile clouds formed by nearby mobile devices and give the mathematical models of the mobile devices and their applications. Secondly, we formulate the scheduling problem in local mobile clouds. After describing the resource discovery scheme and the adaptive, probabilistic scheduling algorithm, we finally validate the performance of the proposed algorithm by simulation experiments.     

A fault‐tolerant dynamic scheduling method on hierarchical mobile edge cloud computing

Mobile edge cloud computing has been a promising computing paradigm, where mobile users could offload their application workloads to low‐latency local edge cloud resources. However, compared with remote public cloud resources, conventional local edge cloud resources are limited in computation capacity, especially when serve large number of mobile applications. To deal with this problem, we present a hierarchical edge cloud architecture to integrate the local edge clouds and public clouds so as to improve the performance and scalability of scheduling problem for mobile applications. Besides, to achieve a trade‐off between the cost and system delay, a fault‐tolerant dynamic resource scheduling method is proposed to address the scheduling problem in mobile edge cloud computing. The optimization problem could be formulated to minimize the application cost with the user‐defined deadline satisfied. Specifically, firstly, a game‐theoretic scheduling mechanism is adopted for resource provisioning and scheduling for multiprovider mobile applications. Then, a mobility‐aware dynamic scheduling strategy is presented to update the scheduling with the consideration of mobility of mobile users. Moreover, a failure recovery mechanism is proposed to deal with the uncertainties during the execution of mobile applications. Finally, experiments are designed and conducted to validate the effectiveness of our proposal. The experimental results show that our method could achieve a trade‐off between the cost and system delay.     

Cost model based service placement in federated hybrid clouds

As cloud federation allows companies in need of computational resources to use computational resources hosted by different cloud providers, it reduces the cost of IT infrastructure by lowering capital and operational expenses. This is the result of economies of scale and the possibility for organizations to purchase just as much computing and storage resources as needed whenever needed. However, a clear specification of cost savings requires a detailed specification of the costs incurred. Although there are some efforts to define cost models for clouds, the need for a comprehensive cost model, which covers all cost factors and types of clouds, is undeniable. In this paper, we cover this gap by suggesting a cost model for the most general form of a cloud, namely federated hybrid clouds. This type of cloud is composed of a private cloud and a number of interoperable public clouds. The proposed cost model is applied within a cost minimization algorithm for making service placement decisions in clouds. We demonstrate the workings of our cost model and service placement algorithm within a specific cloud scenario. Our results show that the service placement algorithm with the cost model minimizes the spending for computational services.     

多编组协同任务分配模型及DLS-QGA 算法求解

为解决多智能体编组协同任务分配问题, 定义任务、智能体编组和相关的分配过程变量, 建立以最高任务执行效率为目标的数学模型. 在问题模型中设计考虑资源损耗的编组资源能力更新机制, 提出用于求解该模型的动态列表规划和量子遗传算法的混合任务分配算法, 使用动态列表规划选择处理的任务, 利用量子遗传算法为选定任务分配最合适编组. 最后通过算例表明, 所提出的方法在解决时序逻辑任务分配时能够得到更优更稳定的方案.

     

A bicriterion approach to time/cost trade-offs in scheduling with convex resource-dependent job processing times and release dates

This paper presents a bicriterion analysis of time/cost trade-offs for the single-machine scheduling problem where both job processing times and release dates are controllable by the allocation of a continuously nonrenewable resource. Using the bicriterion approach, we distinguish between our sequencing criterion, namely the makespan, and the cost criterion, the total resource consumed, in order to construct an efficient time/cost frontier. Although the computational complexity of the problem of constructing this frontier remains an open question, we show that the optimal job sequence is independent of the total resource being used; thereby we were able to reduce the problem to a sequencing one. We suggest an exact dynamic programming algorithm for solving small to medium sizes of the problem, while for large-scale problems we present some heuristic algorithms that turned out to be very efficient. Five different special cases that are solvable by using polynomial time algorithms are also presented.     

Data traffic scheduling algorithm for multiuser OFDM system with adaptive modulation considering fairness among users

Orthogonal frequency division multiplexing (OFDM) is regarded as a very promising digital modulation technique for achieving high rate transmission. However, the increasing number of wireless data users and the deployment of broadband wireless networks have brought about issues of fairness among users and system throughput. In this paper, we propose an efficient scheduling algorithm to maximize system throughput while providing a level of fairness among users for non-real-time data traffic in the downlink of a multiuser OFDM system. We establish a practical scheduling procedure to implement our scheme considering fairness among users and also formulate the resource allocation problem for rate, power, and subcarrier allocation as an integer program that maximizes system throughput. Next, we present a computationally efficient heuristic algorithm for a problem based on the Lagrangian relaxation procedure. Through the computing simulation, we show that the proposed scheme performs better than other schemes in terms of both system throughput and fairness among users.     

Resource allocation optimization for quantitative service differentiation on server clusters

The goal of service differentiation is to provide different service quality levels to meet changing system configuration and resource availability and to satisfy different requirements and expectations of applications and users. In this paper, we investigate the problem of quantitative service differentiation on cluster-based delay-sensitive servers. The goal is to support a system-wide service quality optimization with respect to resource allocation on a computer system while provisioning proportionality fairness to clients. We first propose and promote a square-root proportional differentiation model. Interestingly, both popular delay factors, queueing delay and slowdown, are reciprocally proportional to the allocated resource usage. We formulate the problem of quantitative service differentiation as a generalized resource allocation optimization towards the minimization of system delay, defined as the sum of weighted delay of client requests. We prove that the optimization-based resource allocation scheme essentially provides square-root proportional service differentiation to clients. We then study the problem of service differentiation provisioning from an important relative performance metric, slowdown. We give a closed-form expression of the expected slowdown of a popular heavy-tailed workload model with respect to resource allocation on a server cluster. We design a two-tier resource management framework, which integrates a dispatcher-based node partitioning scheme and a server-based adaptive process allocation scheme. We evaluate the resource allocation framework with different models via extensive simulations. Results show that the square-root proportional model provides service differentiation at a minimum cost of system delay. The two-tier resource allocation framework can provide fine-grained and predictable service differentiation on cluster-based servers.     

Two-machine no-wait flowshop scheduling with learning effect and convex resource-dependent processing times

Two-machine no-wait flowshop scheduling problems in which the processing time of a job is a function of its position in the sequence and its resource allocation are considered in the study. The primary objective is to find the optimal sequence of jobs and the optimal resource allocation separately. Here we propose two separate models: minimizing a cost function of makespan, total completion time, total absolute differences in completion times and total resource cost; minimizing a cost function of makespan, total waiting time, total absolute differences in waiting times and total resource cost. Since each model is strongly NP-hard, we solve both models by breaking them down to two sub-problems, the optimal resource allocation problem for any job sequence and the optimal sequence problem with its optimal resource allocation. Specially, we transform the second sub-problem into the minimum of the bipartite graph optimal matching problem (NP-hard), and solve it by using the classic KM (Kuhn–Munkres) algorithm. The solutions of the two sub-problems demonstrate that the target problems remain polynomial solvable under the proposed model.