Heterogeneous Earliest Finish Time Research Articles

Time-Cost Efficient Scheduling Algorithms for Executing Workflow in Infrastructure as a Service Clouds

Cloud Computing enables delivery of IT resources over the Internet and follows the pay-as-you-go billing model. The cloud infrastructures can be used as an appropriate environment for executing of workflow applications. To execute workflow applications in this environment, it is necessary to develop the workflow scheduling algorithms that consider different QoS parameters such as execution time and cost. Therefore, in this paper we focus on two criteria: total completion time (makespan) and execution cost of workflow, and propose two heuristic algorithms: MTDC (Minimum Time and Decreased Cost) which aims to create a schedule that minimizes the makespan and decreases execution cost, and CTDC (Constrained Time and Decreased Cost) which is based on the first algorithm (MTDC) and aims to create a schedule that decreases the execution cost while satisfying the deadline constraint of the workflow application. The proposed algorithms are evaluated by a simulation process using WorkflowSim. To evaluate the proposed algorithms, the results of MTDC are compared with the results of HEFT (Heterogeneous Earliest Finish Time), and the results of CTDC are compared with the results of heuristic based algorithms [such as IC-PCP (IaaS Cloud Partial Critical Paths), IC-PCPD2 (Deadline Distribution) and BDHEFT (Budget and Deadline HEFT)] and meta-heuristic based algorithms [such as PSO (Particle Swarm Optimization) and CGA2 (Coevolutionary Genetic Algorithm with Adaptive penalty function)]. The results show that the proposed algorithms perform better than the mentioned algorithms in most cases.

PHEFT: Pessimistic Image Processing Workflow Scheduling for DSP Clusters

We address image processing workflow scheduling problems on a multicore digital signal processor cluster. We present an experimental study of scheduling strategies that include task labeling, prioritization, resource selection, and digital signal processor scheduling. We apply these strategies in the context of executing the Ligo and Montage applications. To provide effective guidance in choosing a good strategy, we present a joint analysis of three conflicting goals based on performance degradation. A case study is given, and experimental results demonstrate that a pessimistic scheduling approach provides the best optimization criteria trade-offs. The Pessimistic Heterogeneous Earliest Finish Time scheduling algorithm performs well in different scenarios with a variety of workloads and cluster configurations.

A GSA based hybrid algorithm for bi-objective workflow scheduling in cloud computing

Workflow Scheduling in cloud computing has drawn enormous attention due to its wide application in both scientific and business areas. This is particularly an NP-complete problem. Therefore, many researchers have proposed a number of heuristics as well as meta-heuristic techniques by considering several issues, such as energy conservation, cost and makespan. However, it is still an open area of research as most of the heuristics or meta-heuristics may not fulfill certain optimum criterion and produce near optimal solution. In this paper, we propose a meta-heuristic based algorithm for workflow scheduling that considers minimization of makespan and cost. The proposed algorithm is a hybridization of the popular meta-heuristic, Gravitational Search Algorithm (GSA) and equally popular heuristic, Heterogeneous Earliest Finish Time (HEFT) to schedule workflow applications. We introduce a new factor called cost time equivalence to make the bi-objective optimization more realistic. We consider monetary cost ratio (MCR) and schedule length ratio (SLR) as the performance metrics to compare the performance of the proposed algorithm with existing algorithms. With rigorous experiments over different scientific workflows, we show the effectiveness of the proposed algorithm over standard GSA, Hybrid Genetic Algorithm (HGA) and the HEFT. We validate the results by well-known statistical test, Analysis of Variance (ANOVA). In all the cases, simulation results show that the proposed approach outperforms these algorithms.

Modified HEFT Algorithm for Task Scheduling in Cloud Environment

Cloud computing is now dominated in the area of high performance distributing computing and it provides resource polling and on demand services through internet. Therefore task scheduling becomes an important research area in the field of cloud environment because user’s services demand change dynamically. Heterogeneous Earliest Finish Time (HEFT) unable to distribute the task efficiently. We modify HEFT algorithm that distribute the workload among the processor in effective way and reduce the makespan time of applications. Computational results (Fig. 4-5) shows that modify HEFT algorithm perform better than existing HEFT, Heterogeneous Earliest Finish Time (CPOP) algorithm.

A Deadline-Constrained Multi-Objective Task Scheduling Algorithm in Mobile Cloud Environments

By leveraging the technology of the mobile cloud computing, resource capacity, and computing capability of mobile devices could be extended. However, it is difficult to schedule tasks submitted by mobile users when the number of tasks and service providers increases and to optimize multiple objectives while satisfying users’ requirements. In this paper, the task scheduling is modeled as a multi-objective optimization problem, and we consider both unconstrained and time deadline constrained cases. To address this problem, a heterogeneous earliest finish time (HEFT) using technique for order preference by similarity to an ideal solution method is proposed, which is named as HEFT-T algorithm. For the unconstrained case, a three-stage strategy based on HEFT-T algorithm is presented to select the optimal solutions by applying non-dominated sorting approach. For the deadline-constrained case, an adaptive weight adjustment strategy based on HEFT-T is proposed to adjust weight value for time. Compared with other of the state-of-the-art algorithms, our proposed algorithm performs better in the criterion of both the optimization for total cost as well as mean load, and the deadline-constraint meeting rate.

Green cloud environment by using robust planning algorithm

Cloud computing provided a framework for seamless access to resources through network. Access to resources is quantified through SLA between service providers and users. Service provider tries to best exploit their resources and reduce idle times of the resources. Growing energy concerns further makes the life of service providers miserable. User’s requests are served by allocating users tasks to resources in Clouds and Grid environment through scheduling algorithms and planning algorithms. With only few Planning algorithms in existence rarely planning and scheduling algorithms are differentiated. This paper proposes a robust hybrid planning algorithm, Robust Heterogeneous-Earliest-Finish-Time (RHEFT)1Robust Heterogeneous-Earliest-Finish-Time (RHEFT).1 for binding tasks to VMs. The allocation of tasks to VMs is based on a novel task matching algorithm called Interior Scheduling. The consistent performance of proposed RHEFT algorithm is compared with Heterogeneous-Earliest-Finish-Time (HEFT)2Heterogeneous-Earliest-Finish-Time (HEFT).2 and Distributed HEFT (DHEFT)3Distributed HEFT (DHEFT).3 for various parameters like utilization ratio, makespan, Speed-up and Energy Consumption. RHEFT’s consistent performance against HEFT and DHEFT has established the robustness of the hybrid planning algorithm through rigorous simulations.

A Security Aware Resource Allocation Model for Cloud Based Healthcare Workflows

Objective: Cloud Computing is the new model in distributed computing that provides Software as a service (SaaS), Infrastructure as a service (IaaS), and Platform as a service (PaaS). Researchers utilize the services of cloud for running large scale data and other computation intensive applications like healthcare workflows. The major issue arises while allocating the cloud resources to the task in the workflows, in this paper an efficient scheduling method for work flow has been proposed. Analysis: The applications are modeled as workflows which include dependent tasks that are represented as a Directed Acyclic Graph (DAG) considering every node as a job. The known algorithms cannot solve the problem of resource allocation, and are categorized as NP hard problem. Findings: The proposed hybrid workflow scheduling algorithm which is the combination of Heterogeneous Earliest Finish Time (HEFT) and Security Aware and Budget Aware (SABA) algorithm helps to schedule the workflow efficiently with reduced makespan and required security services.

Deadline‐constrained coevolutionary genetic algorithm for scientific workflow scheduling in cloud computing

SummaryThe cloud infrastructures provide a suitable environment for the execution of large‐scale scientific workflow application. However, it raises new challenges to efficiently allocate resources for the workflow application and also to meet the user's quality of service requirements. In this paper, we propose an adaptive penalty function for the strict constraints compared with other genetic algorithms. Moreover, the coevolution approach is used to adjust the crossover and mutation probability, which is able to accelerate the convergence and prevent the prematurity. We also compare our algorithm with baselines such as Random, particle swarm optimization, Heterogeneous Earliest Finish Time, and genetic algorithm in a WorkflowSim simulator on 4 representative scientific workflows. The results show that it performs better than the other state‐of‐the‐art algorithms in the criterion of both the deadline‐constraint meeting probability and the total execution cost.

Cost-Time Efficient Scheduling Plan for Executing Workflows in the Cloud

The emergence of Cloud Computing as a model of service provisioning in distributed systems instigated researchers to explore its pros and cons on executing different large scale scientific applications, i.e., Workflows. One of the most challenging problems in clouds is to execute workflows while minimizing the execution time as well as cost incurred by using a set of heterogeneous resources over the cloud simultaneously. In this paper, we present, Budget and Deadline Constrained Heuristic based upon Heterogeneous Earliest Finish Time (HEFT) to schedule workflow tasks over the available cloud resources. The proposed heuristic presents a beneficial trade-off between execution time and execution cost under given constraints. The proposed heuristic is evaluated for different synthetic workflow applications by a simulation process and comparison is done with state-of-art algorithm i.e. BHEFT. The simulation results show that our proposed scheduling heuristic can significantly decrease the execution cost while producing makespan as good as the best known scheduling heuristic under the same deadline and budget constraints.

Pareto tradeoff scheduling of workflows on federated commercial Clouds

As distributed computing infrastructures become nowadays ever more complex and heterogeneous, scientists are confronted with multiple competing goals such as makespan in high-performance computing and economic cost in Clouds. Existing approaches typically aim at finding a single tradeoff solution by aggregating or constraining the objectives in an a-priory fashion, which negatively impacts the quality of the solutions. In contrast, Pareto-based approaches aiming to approximate the complete set of (nearly-) optimal tradeoff solutions have been scarcely studied. In this paper, we extend the popular Heterogeneous Earliest Finish Time (HEFT) workflow scheduling heuristic for dealing with multiple conflicting objectives and approximating the Pareto frontier optimal schedules. We evaluate our new algorithm for performance and cost tradeoff optimisation of synthetic and real-world applications in Distributed Computing Infrastructures (DCIs) and federated Clouds and compare it with a state-of-the-art meta-heuristic from the multi-objective optimisation community.

Design and analysis of scheduling strategies for multi-CPU and multi-GPU architectures

We evaluated four scheduling strategies for multi-CPU and multi-GPU architectures.We designed a framework with performance models for task and transfer prediction.Work stealing is efficient with task annotations and data locality heuristics.HEFT cost model performs better on very regular computations. In this paper, we present a comparison of scheduling strategies for heterogeneous multi-CPU and multi-GPU architectures. We designed and evaluated four scheduling strategies on top of XKaapi runtime: work stealing, data-aware work stealing, locality-aware work stealing, and Heterogeneous Earliest-Finish-Time (HEFT). On a heterogeneous architecture with 12 CPUs and 8 GPUs, we analysed our scheduling strategies with four benchmarks: a BLAS-1 AXPY vector operation, a Jacobi 2D iterative computation, and two linear algebra algorithms Cholesky and LU. We conclude that the use of work stealing may be efficient if task annotations are given along with a data locality strategy. Furthermore, our experimental results suggests that HEFT scheduling performs better on applications with very regular computations and low data locality.

Efficient Execution of Microscopy Image Analysis on CPU, GPU, and MIC Equipped Cluster Systems.

High performance computing is experiencing a major paradigm shift with the introduction of accelerators, such as graphics processing units (GPUs) and Intel Xeon Phi (MIC). These processors have made available a tremendous computing power at low cost, and are transforming machines into hybrid systems equipped with CPUs and accelerators. Although these systems can deliver a very high peak performance, making full use of its resources in real-world applications is a complex problem. Most current applications deployed to these machines are still being executed in a single processor, leaving other devices underutilized. In this paper we explore a scenario in which applications are composed of hierarchical data flow tasks which are allocated to nodes of a distributed memory machine in coarse-grain, but each of them may be composed of several finer-grain tasks which can be allocated to different devices within the node. We propose and implement novel performance aware scheduling techniques that can be used to allocate tasks to devices. We evaluate our techniques using a pathology image analysis application used to investigate brain cancer morphology, and our experimental evaluation shows that the proposed scheduling strategies significantly outperforms other efficient scheduling techniques, such as Heterogeneous Earliest Finish Time - HEFT, in cooperative executions using CPUs, GPUs, and MICs. We also experimentally show that our strategies are less sensitive to inaccuracy in the scheduling input data and that the performance gains are maintained as the application scales.

Duplication with Task Assignment in Mesh Distributed System

Load balancing is the major benefit of any distributed system. To facilitate this advantage, task duplication and migration methodologies are employed. As this paper deals with dependent tasks (DAG), we used duplication. Task duplication reduces the overall schedule length of DAG along-with load balancing. This paper proposes a new task duplication algorithm at the time of tasks assignment on various processors. With the intention of conducting proposed algorithm performance computation; simulation has been done on the Netbeans IDE. The mesh topology of a distributed system is simulated at this juncture. For task duplication, overall schedule length of DAG is the main parameter that decides the performance of a proposed duplication algorithm. After obtaining the results we compared our performance with arbitrary task assignment, CAWF and HEFT-TD algorithms. Additionally, we also compared the complexity of the proposed algorithm with the Duplication Based Bottom Up scheduling (DBUS) and Heterogeneous Earliest Finish Time with Task Duplication (HEFT-TD).

A Novel Task Scheduling Algorithm for Heterogeneous Computing

The grid computing system can support the execution of computationally intensive parallel and distributive applications. The main characteristics of grid computing and heterogeneous computing system are similar. A novel scheduling algorithm, called NHEFT is proposed in this paper to enhance the functions of heterogeneous Earliest-Finish time (HEFT) algorithm. The NHEFT algorithm works for a bounded number of heterogeneous processors, the main objective of NHEFT is getting high performance and fast scheduling. The algorithm selects the tasks with a rank system at each step of execution of algorithm, which minimize earliest finish time with the minimization of cost. General Terms Distributed Computing, Heterogeneous Computing, Grid Computing.

Workflow task scheduling algorithm based on resource clustering in cloud computing environment

Focusing on the characteristics of resource under large-scale,heterogeneous and dynamic environment in cloud computing,a workflow task scheduling algorithm based on resource fuzzy clustering was proposed.After quantizing and normalizing the resource characteristics,this algorithm integrated the theory of clustering to divide the resources based on the workflow task model and the resource model constructed in advance.The cluster with better synthetic performance was chosen firstly in scheduling stage.Therefore,it shortened the matching time between the task and the resource,and improved the scheduling performance.By comparing this algorithm with HEFT(Heterogeneous Earliest Finish Time) and DLS(Dynamic Level Scheduling),the experimental results show that the average SLR(Schedule Length Ratio) of this algorithm was smaller than that of HEFT by 3.4%,the DLS by 9.9%,and the average speedup of this algorithm was faster than that of HEFT by 5.9%,the DLS by 10.2% with the increase of tasks in a certain range of [0,100];when the resources were increased in a certain range of [0,100],the average SLR of this algorithm was smaller than that of HEFT by 3.6%,the DLS by 9.7%,and the average speedup of this algorithm was faster than that of HEFT by 4.5%,the DLS by 10.8%.The results indicate that the proposed algorithm realizes the reasonable division of resources,and it surpasses HEFT and DLS algorithms in makespan.

Two-step task scheduling strategy for scientific workflow on cloud computing platform

According to the research and analysis on the existing task scheduling strategy of scientific workflow under the cloud environment,a two-step task scheduling strategy was proposed.This strategy aimed at solving or alleviating the phenomenon of resource idle in Heterogeneous Earliest Finish Time(HEFT) algorithm and SHEFT algorithm.Along with the characteristics of cloud computing environment,it derives from the SHEFT algorithm.It can make the most use of the resources idle time and get the minimum makespan.The experiments and performance analysis for the scheduling strategy show that it has a significant improvement in the workflow makespan and resource utilization.

WEB GRAPH BASED SEARCH BY USING DENSITY OF KEYWORD AND AGE FACTOR

Today, web servers, are the key repositories of the information & internet is the source of getting this information. There is a mammoth data on the Internet. It becomes a difficult job to search out the accordant data. Search Engine plays a vital role in searching the accordant data. A search engine follows these steps: Web crawling by crawler, Indexing by Indexer and Searching by Searcher. Web crawler retrieves information of the web pages by following every link on the site. Which is stored by web search engine then the content of the web page is indexed by the indexer. The main role of indexer is how data can be catch soon as per user requirements. As the client gives a query, Search Engine searches the results corresponding to this query to provide excellent output. Here ambition is to enroot an algorithm for search engine which may response most desirable result as per user requirement. In this a ranking method is used by the search engine to rank the web pages. Various ranking approaches are discussed in literature but in this paper, ranking algorithm is proposed which is based on parent-child relationship. Proposed ranking algorithm is based on priority assignment phase of Heterogeneous Earliest Finish Time (HEFT) Algorithm which is designed for multiprocessor task scheduling. Proposed algorithm works on three on range variable its means the density of keywords, number of successors to the nodes and the age of the web page. Density shows the occurrence of the keyword on the particular web page. Numbers of successors represent the outgoing link to a single web page. Age is the freshness value of the web page. The page which is modified recently is the freshest page and having the smallest age or largest freshness value. Proposed Technique requires that the priorities of each page to be set with the downward rank values & pages are arranged in ascending/ Descending order of their rank values. Experiments show that our algorithm is valuable. After the comparison with Google we find that our Algorithm is performing better. For 70% problems our algorithm is working better than Google.

A PSO-based Genetic Algorithm for Scheduling of Tasks in a Heterogeneous Distributed System

The static task scheduling problem in distributed systems is important because of optimal usage of available machines and accepted computation time for scheduling algorithm. A PSO-based hybrid algorithm is presented to schedule the tasks represented by a Directed Acyclic Graph (DAG) to a bounded number of heterogeneous processors such that its schedule length is optimized. The algorithm first generate feasible initial solutions by using some effective list scheduling strategy and then evolve the solution by using crossover and mutation operator. These operators are modified to ensure the validity of the solutions. And a PSO strategy is also used to improve the quality of the solutions by using local and global information together. The performance of the algorithm is demonstrated by comparing the scheduling length ratio with the existing effectively scheduling algorithms, Heterogeneous Earliest Finish Time and Genetic Scheduling algorithm.

Dynamic task scheduling algorithm with load balancing for heterogeneous computing system

In parallel computation, the scheduling and mapping tasks is considered the most critical problem which needs High Performance Computing (HPC) to solve it by breaking the problem into subtasks and working on those subtasks at the same time. The application sub tasks are assigned to underline machines and ordered for execution according to its proceeding to grantee efficient use of available resources such as minimize execution time and satisfy load balance between processors of the underline machine. The underline infrastructure may be homogeneous or heterogeneous. Homogeneous infrastructure could use the same machines power and performance. While heterogeneous infrastructure include machines differ in its performance, speed, and interconnection. According to work in this paper a new dynamic task scheduling algorithm for Heterogeneous called a Clustering Based HEFT with Duplication (CBHD) have been developed. The CBHD algorithm is considered an amalgamation between the most two important task scheduling in Heterogeneous machine, The Heterogeneous Earliest Finish Time (HEFT) and the Triplet Clustering algorithms. In the CBHD algorithm the duplication is required to improve the performance of algorithm. A comparative study among the developed CBHD, the HEFT, and the Triplet Cluster algorithms has been done. According to the comparative results, it is found that the developed CBHD algorithm satisfies better execution time than both HEFT algorithm and Triplet Cluster algorithm, and in the same time, it achieves the load balancing which considered one of the main performance factors in the dynamic environment.

Implementation of Multi-objective Evolutionary Algorithm for Task Scheduling in Heterogeneous Distributed Systems

This paper presents an effective method for task scheduling in heterogeneous distributed systems. Its objective is to minimize the last task’s finish time and to maximize the system reliability probability. The optimum is carried out through a non-domination sort genetic algorithm. The experimental results based on both randomly generated graphs and the graphs of some real applications showed that, when compared to two well known previous methods, such as heterogeneous earliest finish time (HEFT) algorithm and Critical Path Genetic Algorithm, this algorithm surpasses previous approaches in terms of both last task’s finish time and the system reliability probability.