ArticlePDF Available

Performance Aware Planning Algorithms for Cloud Environments

IGI Global
International Journal of Distributed Systems and Technologies
Authors:
Jyoti Thaman
Jyoti Thaman
Kamal Kumar at Indira Gandhi Delhi Technical University for Women
Kamal Kumar
Download full-text PDF
Read full-text
Download citation

Abstract

For the last decade, cloud computing has been spreading its application base from the small enterprises to the large, from the domestic user to the professional, from buyers to sellers and from research to implementation. Subscribers submit their jobs or workflows for executions on clouds. Workflow scheduling is a very important aspect in cloud computing and it imitates industrial operations, constraints and dependencies. Several approaches such as Greedy, Heuristic, Meta-heuristic and Hybrid have been tried to reschedule workflows. This article proposes Modified HEFT (MHEFT) and Cluster Based Modified HEFT (C-MHEFT). MHEFT modifies the mapping of ranked tasks to the VMs. C-MHEFT is the cluster based extension of MHEFT. The simulations were performed in WorkflowSim and were compared with existing benchmarks in planning algorithms like HEFT and DHEFT. The proposed schemes will help industries, enterprises to model and sequence the Industrial process which will be faster and efficient.
Content uploaded by Kamal Kumar
Author content
All content in this area was uploaded by Kamal Kumar on Jul 27, 2020
Content may be subject to copyright.
DOI: 10.4018/IJDST.2018010101

Volume 9 • Issue 1 • January-March 2018
Copyright © 2018, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
1


Jyoti Thaman, Department of Computer Science and Engineering, Maharishi Markandeshwar University, Mullana,
Ambala, India
Kamal Kumar, University of Petroleum and Energy Studies, Dehradun, India

For the last decade, cloud computing has been spreading its application base from the small enterprises
to the large, from the domestic user to the professional, from buyers to sellers and from research to
implementation. Subscribers submit their jobs or workflows for executions on clouds. Workflow
scheduling is a very important aspect in cloud computing and it imitates industrial operations,
constraints and dependencies. Several approaches such as Greedy, Heuristic, Meta-heuristic and
Hybrid have been tried to reschedule workflows. This article proposes Modified HEFT (MHEFT)
and Cluster Based Modified HEFT (C-MHEFT). MHEFT modifies the mapping of ranked tasks to
the VMs. C-MHEFT is the cluster based extension of MHEFT. The simulations were performed in
WorkflowSim and were compared with existing benchmarks in planning algorithms like HEFT and
DHEFT. The proposed schemes will help industries, enterprises to model and sequence the Industrial
process which will be faster and efficient.

Dynamic Scheduling, HEFT, Performance, Task Types, WorkflowSim

Cloud computing is driving the computing everywhere. A lot of services, applications, infrastructure
and managerial tasks have been mapped on to the cloud platform, both Public and Private. Ever
increasing impact of cloud computing requires a lot of research. It should focus to improve the
performance of clouds. This would prevent drowning of clouds and establish cloud as a long-term
reality. However, increasing pace of cloud platforms has exposed several issues related to QoS
parameters of cloud platforms.
Scheduling of tasks, resource management, network management, privacy and security etc.
should be addressed so that QoS parameters agree with clients and meet to the best possible levels.
Scheduling of tasks is one of the most important challenges as client. Many scheduling and planning
algorithms have been proposed for addressing this issue. HEFT, DHEFT, Min-Min, Max-Min, Round-
Robin, SJF etc. are some well-known proposals. This paper is a proposal of the recently researched
a Modified HEFT algorithm (MHEFT) and Cluster Based Modified HEFT (C-MHEFT). This
MHEFT algorithm is using a novel task-VM mapping based on principle of reducing the difference
between task requirements and VM’s expected performance for given set of tasks. As set of tasks
are changed, VM’s performance gets tuned as per task’s characteristics. This adds to the concept of

Volume 9 • Issue 1 • January-March 2018
2
dynamic self-configuration of VM’s performance when physical characteristics were fixed for whole
length of experimentations. CMHEFT is the extension of MHEFT which modifies the HEFT by
using a clustering approach. In this paper, workflows have been considered as representing different
task types. This proposal has been organized into several sections. Introduction section introduces
the concepts and expectations from the proposal. Related Work presents the existing related work
and identifies the gaps. The Preliminary section presents a briefing of preliminaries in the form of
HEFT planning policy. The fourth section presents proposed Scheduling Approach. The fifth presents
Modified HEFT (MHEFT) planning algorithm and Cluster Based Modified HEFT (C-MHEFT). The
sixth section presents the simulation and analysis of the scheme. Finally comes the Conclusion of
the research and future work.

Recent development and marketing of cloud based solutions have opened up scheduling of task as a
novel and interesting research domain. Several remarkable scheduling algorithms including planning
algorithms have been proposed for task scheduling in literature. This section reviews few remarkable
and citable contributions in research of scheduling of independent and dependents tasks.
Authors Chen and Deelman (2012) have proposed the extension of CloudSim for inclusion
of distributed jobs which are called workflows. The management and scheduling aspects has been
analyzed in a review of workflows application (Liu et al., 2014). A useful conclusion was drawn from
the review that single cloud model is inefficient with respect to increasing scale and complexity of
workflows. Another proposal (Zhou & Garg, 2015) evaluated the performance of various existing
workflow scheduling schemes for different workflows. This work has endorsed that dynamism may
be important parameter to keep in knowledge while evaluating any workflow scheduling proposal.
Some study (Thaman & Singh, 2016a) has proposed a Robust HEFT (RHEFT) planning algorithm
by considering a novel VM-Task mapping algorithm. Authors Thaman and Singh (2016b) have
presented a simulated performance of various dynamic scheduling schemes available in WorkflowSim.
Authors (Chopra & Singh, 2013) have proposed hybrid of HEFT with a novel concept of sub-
deadline to achieve deadline constraints. Almost all the tasks were scheduled within their deadline.
A novel mechanism called Partial Deadline Constraint (PDC) was proposed which not only achieves
a better mapping but also exploits the underlying concept of elasticity (Arabnejad & Bubendorfer,
2015). Authors Thaman and Singh (2016c) have proposed a novel taxonomy for classification of
task scheduling algorithms based on the goal oriented and constraint oriented approach. This paper
presents a highly useful current perspective of scheduling approaches. Another research (Singh et al.,
2014) has presented a comparison of various scheduling algorithms on distinct aspects like suitability,
feasibility and adaptability to cloud platform. This paper also proposed a hybrid scheduling approach
which can bring better returns to cloud service providers and ensure better quality of services. A
Genetic Algorithm based load balancing approach was presented (Dasgupta et al., 2013). Authors have
discussed and proved that dynamic nature of cloud environment might be the best addressed through
adaptive scheduling schemes. The proposal claimed that un-balanced state is inevitable whereas
system might have started well in balanced state. Another Genetic Algorithm based heuristic was
proposed (Yu & Buyya, 2006) where two objectives, namely budget and deadlines were addressed.
Optimizations were achieved by either minimizing execution times while at the same time meeting
the budget and deadline constraints posed by the subscriber. Evaluation was done using both balanced
and unbalanced workflows and results show this newly proposed approach was better in handling
complex workflow structures. Authors (Sheng & Li, 2016) have proposed a new concept named
Template Based Genetic Algorithm (TBGA) where users specify QoS constraints. Based on specified
values, TBGA computes the maximal number of tasks that could be scheduled on each machine.
These computed results are known as template. These templates are used to segregate the tasks into
multiple subsets which are to be allocated to one processor each. This allocation was made considering
13 more pages are available in the full version of this
document, which may be purchased using the "Add to Cart"
button on the product's webpage:
www.igi-global.com/article/performance-aware-planning-
algorithms-for-cloud-environments/196264?camid=4v1
This title is available in InfoSci-Computer Systems and
Software Engineering eJournal Collection, InfoSci-Journals,
InfoSci-Journal Disciplines Computer Science, Security, and
Information Technology, InfoSci-Journal Disciplines
Engineering, Natural, and Physical Science, InfoSci-Select,
InfoSci-Computer Science and IT Knowledge Solutions –
Journals. Recommend this product to your librarian:
www.igi-global.com/e-resources/library-
recommendation/?id=163
Related Content
A Congestion Controlled and Load Balanced Selection Strategy for Networks
on Chip
Ashima Arora and Neeraj Kumar Shukla (2020). International Journal of Distributed
Systems and Technologies (pp. 1-14).
www.igi-global.com/article/a-congestion-controlled-and-load-balanced-
selection-strategy-for-networks-on-chip/240772?camid=4v1a
Failure Detector of Perfect P Class for Synchronous Hierarchical Distributed
Systems
Anshul Verma and K. K. Pattanaik (2016). International Journal of Distributed
Systems and Technologies (pp. 57-74).
www.igi-global.com/article/failure-detector-of-perfect-p-class-for-
synchronous-hierarchical-distributed-systems/151579?camid=4v1a
Cost Evaluation on Building and Operating Cloud Platform
Yue-Shan Chang, Yi-Kang Lee, Tong-Ying Juang and Jing-Shyang Yen (2013).
International Journal of Grid and High Performance Computing (pp. 43-53).
www.igi-global.com/article/cost-evaluation-on-building-and-operating-cloud-
platform/78895?camid=4v1a
A Fuzzy Real Option Model to Price Grid Compute Resources
David Allenotor, Ruppa K. Thulasiram, Kenneth Chiu and Sameer Tilak (2010).
Handbook of Research on Scalable Computing Technologies (pp. 471-485).
www.igi-global.com/chapter/fuzzy-real-option-model-
price/36421?camid=4v1a
... Workflow's task allocation is one of the most common application models, particularly in designing, business plans, and logical fields which has been directed towards workflow task allocation. Therefore, many researchers address the workflow tasks allocation problem for many aspects such as computational time minimization [9][10][11][12][13][14][15], minimum budget assurance [16,17], least energy consumption [18,19], improving the throughput and server performance [20,21] and employing security measures [22][23][24][25][26][27][28][29][30][31][32] in heterogeneous and other efficient computing [33,34] for single workflow. Further, the work has been also reported for multiple workflows allocation to improve the turnaround time, response time and flowtime [35][36][37][38][39][40][41][42][43], budget [43,44], and energy [45,46], by aggregating the batch of workflows prior to allocation. ...
Security prioritized multiple workflow allocation model under precedence constraints in cloud computing environment
Article
Full-text available
  • Jan 2023
  • CLUSTER COMPUT
In the last decade, cloud computing has become an effective and efficient service delivery platform to offer resources, innovation, and economies of scale on telecommunication networks. An allocation scheme without security constraints consideration may assign the high security-sensitive tasks onto the lower trustworthy machines. This would lead the performance deterioration. To address this issue, in this paper, a security-prioritized multiple workflow allocation (SPMWA) model is proposed by integrating the security-prioritized mapping scheme for Infrastructure-as-a-Service cloud computing environment. It is expected that incorporating a security-prioritized allocation scheme under precedence constraints will enhance the performance of workflow processing in risk-prone environments. In this model, more priority is given to tasks with high-security demand to get allocated onto the more trustworthy virtual machines during allocation to minimize the failure probability of the cloud system. The failure probability can be minimized by assigning the tasks on to the trustable virtual machines exhibiting sufficient trust level to minimize the number of task failures. The number of task failure, failure probability, and makespan have been computed for the comparative evaluation of the SPMWA. For performance comparison, the SPMWA model has been compared with state-of-the-art models from the literature having the same environment and objective. The experimental evaluation confirms the superior performance of the proposed model on account of the considered objective among its peers.
View
GMG-NCDVAE: Guided de novo Molecule Generation using NLP Techniques and Constrained Diverse Variational Autoencoder
Article
  • Aug 2023
Text processing techniques in Natural Language Processing (NLP) find applications in many industries such as pharmaceutical, automation, and automotive. Drug design using variational autoencoders is a popular data-assisted technique to design drug molecules with control over molecular properties. It generates continuous latent space, which can be optimized. This paper introduces a constrained variational autoencoder-based molecular generation structure using the SMILES format. The proposal is accompanied by the generation of molecules, filtering them based on scores, and subsequently determining the optimal molecules by using NLP matured techniques. To generate more meaningful latent space, a condition vector of molecular properties is combined with the SMILES representation of molecules. A tunable parameter (diversity,D) is also used to control the diversity in the generated molecules. The proposed architecture is evaluated using standard datasets. Validity, uniqueness, and FCD are evaluation matrices used to access the performance of model. The validity of proposed model is maximum (92.11%) at diversity level 1. As diversity level increases the validity of generated molecules decreases. This is intuitively consistent because increased diversity reduces replicas and improves variety in the generated molecules. Thus proposed model provide control over diversity of generated molecules. The results clearly indicate that the proposed method outperforms other SMILE based methods and gives a new direction for the generation of desired molecules.
View
Improving Virtual Machine Migration Effects in Cloud Computing Environments Using Depth First Inspired Opportunity Exploration
Article
Full-text available
  • Jan 2022
The cloud platform has established itself as the de-facto standard in IT outsourcing. This is resulting in large-scale migration of infrastructure and development platforms from in-house to cloud service providers. Many recent proposals on cloud platforms have addressed several issues that appeared on the cloud horizon. VM placement (VMP) has been a serious concern when it comes to placement of VMs after migration or VM reallocation. Most of the recent works have lacked multiple VM placement (MVMP) problem instances. A recently researched idea of MVMP through depth first opportunistic exploration (DFOE) is proposed in this paper. The performance of MVMP is compared with existing single VM placement benchmark algorithm. Improvement in terms of number of VM migrations, energy consumption, and VM reallocation is reported through simulation of real-time load scenario. Cloud environments can benefit from MVMP and improve operating margins in terms of power saving and load balancing.
View
Opportunistic Two Virtual Machines Placements in Distributed Cloud Environment
Article
Full-text available
  • Oct 2020
Cloud computing is a potentially tremendous platform and its presence is experienced in day to day life. Most infrastructure and technology enterprises have migrated to a cloud-based infrastructure and storage. With so much dependence on the cloud as a distributed and reliable platform, but a few issues remain as a challenge and provide food for the ever-active research entity. Considering a very basic aspect of VM migration followed by VM placement, one VM at a time is a prominent approach. This article presents a novel idea of placing two VMs at a time. This proposal is a draft of solution for the Two VM Placement problem. The experimental validation was done against a well-known placement algorithm, the power aware best fit decreasing (PABFD). PABFD and TVMP were applied on a given context and results were obtained for three important parameters, which include the number of VM migrations, reallocation means, and energy efficiency. Improvements on these parameters may prove beneficial.
View
Optimization of Task Scheduling in Cloud Computing by MOGA: Cloud Computing
Book
Full-text available
  • May 2017
This book is about a guide for using Cloud Computing System. In this book, we’ll explore the value of collaborative approaches that emphasize equal responsibility over hierarchy. Nowadays availability of high capacity networks low cost computers and storage devices as well as the global adoption of hardware virtualization service-oriented architecture and autonomic and utility computing have guide to a growth in cloud computing. Cloud computing has become a highly demanded service or utility due to the advantages of fast computing power, low cost of services, high performance, scalability, accessibility as well as availability. But in cloud computing task scheduling and provision of resources are main problem areas. Therefore the research work is motivated to drive the task scheduling based on multi-objective genetic algorithm employs an improved cost-based scheduling algorithm for making an efficient mapping of tasks to available resources in cloud computing environment. This book specifies inclusive knowledge of the most meaningful characteristics of Cloud Computing Systems.
View
Current Perspective in Task Scheduling Techniques in Cloud Computing: A Review
Article
Full-text available
  • Jan 2016
Cloud computing is a development of parallel, distributed and grid computing which provides computing potential as a service to clients rather than a product. Clients can access software resources, valuable information and hardware devices as a subscribed and monitored service over a network through cloud computing.Due to large number of requests for access to resources and service level agreements between cloud service providers and clients, few burning issues in cloud environment like QoS, Power, Privacy and Security, VM Migration, Resource Allocation and Scheduling need attention of research community.Resource allocation among multiple clients has to be ensured as per service level agreements. Several techniques have been invented and tested by research community for generation of optimal schedules in cloud computing. A few promising approaches like Metaheuristics, Greedy, Heuristic technique and Genetic are applied for task scheduling in several parallel and distributed systems. This paper presents a review on scheduling proposals in cloud environment.
View
Cost-effective task scheduling using hybrid approach in cloud
Article
  • Jan 2017
Cloud computing is the most recent computing paradigm where IT services are provided and delivered over the internet on demand. With increasing demand for cloud services models, some new challenges in cloud computing are also gaining importance. Increasing energy concern has not even spared cloud environment for considering energy conscious solutions. The scheduling problem for cloud environment has a lot of awareness as the cloud subscriber's tasks could be mapped to the available resources to achieve better results. Traditional heuristics are popularly used to obtain schedules for given tasks and resources. Best efforts results from these heuristics depend upon the tasks and resources at disposal. Meta-heuristic-based scheduling solutions are iterationbased solutions. Meta-heuristic performs search over solution space and converges in minimum number of iterations. A novel heuristic Nearest Neighbour (NN) and a hybrid variant of Particle Swarm Optimisation (PSO) as Nearest Neighbour Cost-Aware PSO (NNCA-PSO) have been proposed and implemented in this paper. Performance of NNCA-PSO has been evaluated for various parameters like makespan, energy consumption, utilisation and economic efficiency. NNCA-PSO has been proved to be stable like PSO and finds near-optimal solutions.
View
View
View
Dynamic Fault Tolerant Scheduling Mechanism for Real Time Tasks in cloud computing
Conference Paper
  • Mar 2016
Now-a-days cloud makes a success in various fields such as financial transaction, scientific computing and so on. Reliability and Availability are most important between cloud provider and user. Fault tolerance plays a major role for using cloud services for industry and scientific specific purposes. In the present work, we focus on fault tolerance and resource allocation policies for cloud environments. In the existing system, Primary Backup (PB) model is used, but it does not contain any dynamic resource allocating mechanism. We propose a dynamic resource allocating mechanism with fault tolerance to improve resource utilization. We incorporate a backup overlapping mechanism and efficient VM migration strategy for designing novel Dynamic Fault Tolerant Scheduling Mechanism for Real Time Tasks in cloud computing. Our proposed model aims at achieving both fault tolerance and high resource utilization in the cloud. The experiments are evaluated using random synthetic workload and Google cloud trace logs to test the efficiency of our proposed model.
View
View
View
View
Extending Dynamic Scheduling Policies in WorkflowSim by Using Variance based Approach
Article
  • Apr 2016
Workflow scheduling has been around for more than two decades. With growing interest in service oriented computing architecture among researchers and corporate users, different platform like clusters computing, grid computing and most recent cloud computing, appeared on computing horizon. Cloud computing has attracted a lot of interest from all types of users. It gave rise to variety of applications and tasks with varied requirements. Heterogeneity in application's requirement catalyzes the provision of customized services for task types. Representation of tasks characteristics and intertask relationship through workflows is in use since the ages of automation. Scheduling of workflows not only maintain the hierarchical relationship between the tasks but also dictates the requirement of dynamic scheduling. This paper presents a variance based extensions of few promising dynamic scheduling policies supported by WorkflowSim. An exhaustive performance analysis presents strength and weakness of the authors' proposal. © Copyright 2016, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
View