Skip to main content

suvendu nayak

Followers

5

Following

5

Public Views

Researcher in computer science
Address: India

less

University of California, Santa Barbara

Universiti Teknologi Malaysia - UTM

B. Harun Küçük

University of Pennsylvania

Alysson M. Costa

Universidade de São Paulo

University of Haifa

Université de Genève

Armando Marques-Guedes

UNL - New University of Lisbon

Peter Schaeffer

West Virginia University

Dublin Institute of Technology

Qusay F. Hassan

InterestsView All (13)

Uploads

Papers by suvendu nayak

Study and impact analysis of COVID-19 pandemic clinical data on infection spreading

Elsevier eBooks, 2022

In December 2019, a severe pneumonialike disease has occurred in the city of Wuhan, Hubei Provinc... more In December 2019, a severe pneumonialike disease has occurred in the city of Wuhan, Hubei Province in China. Within a very short period the infection spread across the whole world, but there was no previous medical history about this virus and how, where, and when the disease infected the human body and mutated in humans is still unknown. Subsequently, the coronavirus disease 2019 (COVID-19) outbreak was declared as the world pandemic on March 2020 by the World Health Organization because of its harmfulness and super spreading nature. Till now, there is no specific medications and clinical treatment available to avoid this pandemic COVID-19 outbreak. For this, it is essential to have a detailed study and analysis through the recent technologies. The recent trends such as artificial intelligence and machine learning (ML) based models can learn from past patient medication data and can suggest improvement accordingly by analyzing the data to control the spread. In the present scenario, the correct decision could be the appropriate precaution to stop spreading as well as controlling such a pandemic disease by proposing predictive ML that analyzes past data and conclude some useful information for future control of the spread of COVID-19 infections using minimum resources. The ML-based approach can be helpful to design different models to give a predictive solution for controlling infection and spreading and taking precaution toward the COVID-19 outbreak. In this chapter, we study the basic information of COVID-19 and its effectiveness worldwide. We also state the fundamental steps of ML, discuss the ML mechanism to study the pandemic for research and academic purposes, and study the data analytics of clinical data of India through a case study. As the data is a time series data, we analyze the data from March 1, 2020 to April 15, 2020; the decision tree approach of ML is discussed through a case study. Finally, the chapter is concluded with certain future scope of work in this area of research.

A Deep Learning-based Compression and Classification Technique for Whole Slide Histopathology Images

arXiv (Cornell University), May 11, 2023

Design Model of Smart “Anganwadi Center” for Health Monitoring

We live in an Era where the technology has taken over all the sphere of our daily life, but still... more We live in an Era where the technology has taken over all the sphere of our daily life, but still, the World Bank estimates that India is one of the highest-ranking countries in the world for the number of children suffering from malnutrition. It is mostly seen in rural areas. Some of the infants die due to hunger, malnutrition and related disease. We know doctors in rural areas visits at most twice a month but it is not sufficient for proper health care therefore somehow if we narrow-down he gap of regular checkups then proper health care could be seen in rural areas also. In this paper, we make a portable device (wearable gear) which can be used to measure different health parameters and proper care of child can be done accordingly. The chapter brings an attempt and interest among the researchers to monitor the health of rural child in different “Anganwadi centers”. We have implemented the working model and study its different performance parameters.

Weighted Clustering Ensemble with Base Clustering Frequency and Diversity

International journal of electronics engineering and application, Sep 30, 2021

Clustering Ensemble, also referred as Consensus Clustering, is a tool for enhancing the reliabili... more Clustering Ensemble, also referred as Consensus Clustering, is a tool for enhancing the reliability and stability of data clustering by aggregating the base clusterings obtained by different clustering algorithms in the input ensemble. This study introduces a novel ensemble selection strategy for establishing consensus clustering. Our strategy avoids looking at the entire population of base clusterings in the ensemble in order to establish a quality consensus by carefully selecting a few base clusterings. The experimental results reveal that the suggested method's consensus clustering surpasses some other well-known clustering ensemble methods in terms of clustering accuracy for diverse data sets.

Analyzing the Impact of Adversarial Examples on Explainable Machine Learning

arXiv (Cornell University), Jul 17, 2023

Adversarial attacks are a type of attack on machine learning models where an attacker deliberatel... more Adversarial attacks are a type of attack on machine learning models where an attacker deliberately modifies the inputs to cause the model to make incorrect predictions. Adversarial attacks can have serious consequences, particularly in applications such as autonomous vehicles, medical diagnosis, and security systems. Work on the vulnerability of deep learning models to adversarial attacks has shown that it is very easy to make samples that make a model predict things that it doesn't want to. In this work, we analyze the impact of model interpretability due to adversarial attacks on text classification problems. We develop an ML-based classification model for text data. Then, we introduce the adversarial perturbations on the text data to understand the classification performance after the attack. Subsequently, we analyze and interpret the model's explainability before and after the attack.

JOB-DCA: A Cost Minimizing Jaya Optimization-Based Data Center Allocation Policy for IaaS Cloud Model

Advances in intelligent systems and computing, 2021

Multicriteria decision-making techniques for avoiding similar task scheduling conflict in cloud computing

International Journal of Communication Systems, Aug 16, 2019

This is a PDF file of an article that has undergone enhancements after acceptance, such as the ad... more This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

An enhanced deadline constraint based task scheduling mechanism for cloud environment

Journal of King Saud University - Computer and Information Sciences, Feb 1, 2022

In this work, we present a scheduling mechanism for deadline sensitive task or lease. It is very ... more In this work, we present a scheduling mechanism for deadline sensitive task or lease. It is very impelling and challenging to allocate on-demand resources within the deadline rather than rejecting the lease. The work studies the insight of existing backfilling algorithm that is used to schedule deadline sensitive lease in the Open Nebula cloud platform. In backfilling mechanism a lease is selected in First Come First Serve (FCFS) to be backfilled in which some ideal resources can be found out and allocated to other leases. Mainly, the scheduling performance of backfilling algorithm could be better when there are conflicts among the similar leases to backfill. Moreover, backfilling algorithm doesn't allow to schedule a new deadline sensitive task during execution. The proposed work addresses such types of issues and improves the scheduling performances with respect to maximize the task acceptance ratio and reduce the task rejection ratio. Finally, the performances are compared with the existing mechanisms that are quite good.

Offer Based Auction Mechanism for Virtual Machine Allocation in Cloud Environment

Advances in intelligent systems and computing, 2020

Nowadays, the cloud service providers provide different resources for computation such as CPU pow... more Nowadays, the cloud service providers provide different resources for computation such as CPU power, memory to a wide range of users by virtual machines (VM). The VMs vary according to their capacity of resources, where each has a limited number of resources with a different price. For this, it is very much required to allocate the available resources efficiently among the requested users, So that there should be the minimization of the cost of resources and maximization of profit from the resources. Therefore, resource allocation is one of the most challenging issues in cloud computing and fog computing. Presently, in cloud computing VMs are allocated with a fixed price mechanism which is not minimizing the cost. In this paper, to solve such problems, we purposed a hybrid combinatorial auction-based approach for allocation of VMs. Here, we calculate the cost with maximum resource utilization which is beneficiary to users on demand.

Task Scheduling Mechanism Using Multi-criteria Decision-making Technique, MACBETH in Cloud Computing

Advances in intelligent systems and computing, 2018

Cloud computing is the new era of Internet technology which provides various utilities and comput... more Cloud computing is the new era of Internet technology which provides various utilities and computing resources from the pool of resources on the basis of “pay per Use”. It is challenging one to allocate required on-demand resources for all the users’ request. Meanwhile, the service provider aims toward a better resource utilization. These user requests are called task, if task execution is bounded by time limit which is called deadline-based task. The deadline-based tasks have different parameters. To schedule these tasks, researchers proposed many works based upon these parameters. However, in this work, we considered the scheduling of deadline-based task that is a Multi-criteria Decision-making problem due to different task’s parameters associated with it. The work is proposed to implement Measuring Attractiveness through a Category-Based Evaluation Technique (MACBETH) to ranking the deadline-based task by which many tasks can meet their deadline. The results of the proposed work are quite good as compared to the existing mechanisms.

Deadline sensitive lease scheduling in cloud computing environment using AHP

Journal of King Saud University - Computer and Information Sciences, Apr 1, 2018

The OpenNebula is an open source environment which provides cloud resources with the help of Haiz... more The OpenNebula is an open source environment which provides cloud resources with the help of Haizea as a lease manager. The Haizea supports different types of leases from which deadline sensitive lease is one of them. In real time, most of the leases are deadline sensitive leases. These deadline sensitive leases are scheduled by using the backfilling algorithm. In the backfilling algorithm one of the lease is selected from the best effort queue which will provide the free resources to schedule the deadline sensitive lease. But in some scenario backfilling algorithm does not provide better scheduling if there is similar types of leases and must be in conjugative in sequence. This work aims to use AHP (Analytic Hierarchy Process) as a decision maker in the backfilling algorithm to choose the possible best lease from the given best effort queue in order to schedule the deadline sensitive lease. The proposed work improves the performance of the backfilling algorithm by scheduling more number of leases and minimizing the lease rejection using AHP.

An algorithm that earning users' trust on cloud

Now cloud computing provides many opportunities for enterprises by offering a range of computing ... more Now cloud computing provides many opportunities for enterprises by offering a range of computing services. Cloud computing has opened up a new frontier of challenges by introducing a different type of trust scenario. Today, the problem of trusting cloud computing is a paramount concern for most enterprises. It's not that the enterprises don't trust the cloud providers' intentions rather, they question cloud computing's capabilities regarding the security. To trust on the cloud, the cloud computing should have the appropriate technologies to tackle the accompanying challenges of trust. In this paper we proposed an algorithm which can increase the trust of the client on the cloud service provider. Different researchers have proposed different algorithms like data encryption and data decryption using private or public key concepts. But trust to cloud is basically based on cloud trust policies, which is not enough for the client to use cloud for the sensitive data. In this paper the proposed algorithm uses a function called key_function and the key which increases the trust to cloud during transmission. The cloud can ensure that the cloud server is providing necessary security for the user's data during transmission. By matching the key which is updated by the key_function, the user can be secured that the data accessed from the cloud server is secured.

An Improved Task Scheduling Mechanism Using Multi-Criteria Decision Making in Cloud Computing

International Journal of Information Technology and Web Engineering, Apr 1, 2019

Inthiswork,theauthorsproposeMulti-criteriaDecision-makingtoscheduledeadlinebasedtasksin cloudcomp... more Inthiswork,theauthorsproposeMulti-criteriaDecision-makingtoscheduledeadlinebasedtasksin cloudcomputing.Theexistingbackfillingtaskschedulingalgorithmcouldnothandlesimilartasks forscheduling.Inbackfillingalgorithm,tasksarebackfilledtoprovideidealresourcestoschedule otherdeadlinesensitivetasks.However,thetasktobebackfilledisselectedonfirstcome,firstserve (FCFS)basisfromschedulingqueue.Theschedulingperformancesrequiretobeimprovedwhen, therearesimilartasks.Inthisproposedwork,theauthorsproposetoimplementMCDMtechnique, TOPSIS(TechniqueforOrderPreferencebySimilaritytoIdealSolution)toimprovetheperformance of the backfilling algorithm through scheduling deadline sensitive tasks in cloud computing. It resolvestheconflictsamongthesimilartasksthatisusedasadecisionsupportsystem.Theworkis simulatedwithsyntheticdatasetsbasedonslackvaluesofthetasks.Theperformanceresultsaffirm thetaskcompletionandreductionintaskrejectioncomparedtotheexistingbackfillingalgorithm.

Truthful Resource Allocation Detection Mechanism for Cloud Computing

Resource allocation is a n-p hard problem. In spite of different proposed algorithms and methods,... more Resource allocation is a n-p hard problem. In spite of different proposed algorithms and methods, still it is a challenging one. VMs are basically considered as a resource in cloud computing. In his paper, we are considering the resource allocation in Haizea model which is an open-source VM-based lease management architecture and acts as a resource manager. Different scheduling mechanisms are used in Haizea like Advance Reservation (AR), Best effort (BE), Immediate (IM) and Deadline Sensitive (DS). Except these scheduling swapping and backfilling algorithm is proposed. But Backfilling is not applicable for all types of situation. In this work we tried to make a truthful scheduling by our proposed mechanism. The proposed mechanism also identifies where the lease can be scheduled or not. The mechanism for allocation detection is simple and robust.

A new method for weighted ensemble clustering and coupled ensemble selection

Connection science, Jan 7, 2021

ABSTRACT Clustering ensemble, also referred to as consensus clustering, has emerged as a method o... more ABSTRACT Clustering ensemble, also referred to as consensus clustering, has emerged as a method of combining an ensemble of different clusterings to derive a final clustering that is of better quality and robust than any single clustering in the ensemble. Normally clustering ensemble algorithms in the literature combine all the clusterings without learning the ensemble. But by learning the ensemble, one can define the merit of a clustering or even a cluster in it, in forming a quality consensus. In this work, we propose a cluster-level surprisal measure to define the merit of a clustering that reflects both levels of agreement as well as disagreement among clusters. Using the proposed measure of merit, we devise a polynomial heuristics that judiciously selects a subset of clusterings from the ensemble that contribute positively in forming the consensus. We also empirically show that consensus achieved by our proposed method performs better in terms of quality compared to well-known clustering ensemble algorithms on different benchmark datasets.

Modeling of Task Scheduling Algorithm Using Petri-Net in Cloud Computing

Advances in intelligent systems and computing, 2018

Task scheduling is an NP-hard problem. In earlier, different task scheduling algorithms are propo... more Task scheduling is an NP-hard problem. In earlier, different task scheduling algorithms are proposed for cloud computing environment. By adapting new and modified task scheduling algorithms, better resource utilization can be obtained. In real time, most of the tasks are deadline-based tasks. The deadline-based task has different parameters. The backfilling algorithm is used to schedule these types of tasks in Haizea. In this paper, we modeled the existing backfilling algorithm for scheduling deadline-based task using Petri-Net. The paper presents the design model of the existing backfilling algorithm. The model specifies real-time challenges of backfilling algorithm using Petri-Net. The work also comes forward with some design issues of backfilling algorithm using Petri-Net.

Petri Net: Design and Analysis of Parallel Task Scheduling Algorithm

Lecture notes in electrical engineering, Oct 29, 2017

In real time, most of the tasks are deadline based. The deadline-based task has different paramet... more In real time, most of the tasks are deadline based. The deadline-based task has different parameters: arrival time, start time, execution time, and deadline. Many performance-based task scheduling algorithms are proposed by number of researchers theoretically. But due to change of implement environment, the performance varies. Petri net is a graphical and mathematical model to evaluate and analysis of the system. In Petri net, conflicts are occurred during firing. In this paper, we designed and modeled the Petri net for scheduling deadline-based task by resolving the conflicts. We also proposed a scheduling mechanism and firing rules to schedule deadline-based tasks. The designed model increases the resource utilization of a physical system in cloud computing. The performance of the proposed model is analyzed using the PIPE v4.3.0. We analyzed the reach ability graph, convertibility graph, and steady-state analysis of the model.

Energy efficient task allocation and consolidation in multicast cloud network

Wireless Networks

Renewable energy source based quality of service (QoS)-aware routing mechanism in cloud network

Wireless Networks

JOB-DCA: A Cost Minimizing Jaya Optimization-Based Data Center Allocation Policy for IaaS Cloud Model

Advances in Intelligent Systems and Computing, 2021

Study and impact analysis of COVID-19 pandemic clinical data on infection spreading

Elsevier eBooks, 2022

In December 2019, a severe pneumonialike disease has occurred in the city of Wuhan, Hubei Provinc... more In December 2019, a severe pneumonialike disease has occurred in the city of Wuhan, Hubei Province in China. Within a very short period the infection spread across the whole world, but there was no previous medical history about this virus and how, where, and when the disease infected the human body and mutated in humans is still unknown. Subsequently, the coronavirus disease 2019 (COVID-19) outbreak was declared as the world pandemic on March 2020 by the World Health Organization because of its harmfulness and super spreading nature. Till now, there is no specific medications and clinical treatment available to avoid this pandemic COVID-19 outbreak. For this, it is essential to have a detailed study and analysis through the recent technologies. The recent trends such as artificial intelligence and machine learning (ML) based models can learn from past patient medication data and can suggest improvement accordingly by analyzing the data to control the spread. In the present scenario, the correct decision could be the appropriate precaution to stop spreading as well as controlling such a pandemic disease by proposing predictive ML that analyzes past data and conclude some useful information for future control of the spread of COVID-19 infections using minimum resources. The ML-based approach can be helpful to design different models to give a predictive solution for controlling infection and spreading and taking precaution toward the COVID-19 outbreak. In this chapter, we study the basic information of COVID-19 and its effectiveness worldwide. We also state the fundamental steps of ML, discuss the ML mechanism to study the pandemic for research and academic purposes, and study the data analytics of clinical data of India through a case study. As the data is a time series data, we analyze the data from March 1, 2020 to April 15, 2020; the decision tree approach of ML is discussed through a case study. Finally, the chapter is concluded with certain future scope of work in this area of research.

A Deep Learning-based Compression and Classification Technique for Whole Slide Histopathology Images

arXiv (Cornell University), May 11, 2023

Design Model of Smart “Anganwadi Center” for Health Monitoring

We live in an Era where the technology has taken over all the sphere of our daily life, but still... more We live in an Era where the technology has taken over all the sphere of our daily life, but still, the World Bank estimates that India is one of the highest-ranking countries in the world for the number of children suffering from malnutrition. It is mostly seen in rural areas. Some of the infants die due to hunger, malnutrition and related disease. We know doctors in rural areas visits at most twice a month but it is not sufficient for proper health care therefore somehow if we narrow-down he gap of regular checkups then proper health care could be seen in rural areas also. In this paper, we make a portable device (wearable gear) which can be used to measure different health parameters and proper care of child can be done accordingly. The chapter brings an attempt and interest among the researchers to monitor the health of rural child in different “Anganwadi centers”. We have implemented the working model and study its different performance parameters.

Weighted Clustering Ensemble with Base Clustering Frequency and Diversity

International journal of electronics engineering and application, Sep 30, 2021

Clustering Ensemble, also referred as Consensus Clustering, is a tool for enhancing the reliabili... more Clustering Ensemble, also referred as Consensus Clustering, is a tool for enhancing the reliability and stability of data clustering by aggregating the base clusterings obtained by different clustering algorithms in the input ensemble. This study introduces a novel ensemble selection strategy for establishing consensus clustering. Our strategy avoids looking at the entire population of base clusterings in the ensemble in order to establish a quality consensus by carefully selecting a few base clusterings. The experimental results reveal that the suggested method's consensus clustering surpasses some other well-known clustering ensemble methods in terms of clustering accuracy for diverse data sets.

Analyzing the Impact of Adversarial Examples on Explainable Machine Learning

arXiv (Cornell University), Jul 17, 2023

Adversarial attacks are a type of attack on machine learning models where an attacker deliberatel... more Adversarial attacks are a type of attack on machine learning models where an attacker deliberately modifies the inputs to cause the model to make incorrect predictions. Adversarial attacks can have serious consequences, particularly in applications such as autonomous vehicles, medical diagnosis, and security systems. Work on the vulnerability of deep learning models to adversarial attacks has shown that it is very easy to make samples that make a model predict things that it doesn't want to. In this work, we analyze the impact of model interpretability due to adversarial attacks on text classification problems. We develop an ML-based classification model for text data. Then, we introduce the adversarial perturbations on the text data to understand the classification performance after the attack. Subsequently, we analyze and interpret the model's explainability before and after the attack.

JOB-DCA: A Cost Minimizing Jaya Optimization-Based Data Center Allocation Policy for IaaS Cloud Model

Advances in intelligent systems and computing, 2021

Multicriteria decision-making techniques for avoiding similar task scheduling conflict in cloud computing

International Journal of Communication Systems, Aug 16, 2019

This is a PDF file of an article that has undergone enhancements after acceptance, such as the ad... more This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

An enhanced deadline constraint based task scheduling mechanism for cloud environment

Journal of King Saud University - Computer and Information Sciences, Feb 1, 2022

In this work, we present a scheduling mechanism for deadline sensitive task or lease. It is very ... more In this work, we present a scheduling mechanism for deadline sensitive task or lease. It is very impelling and challenging to allocate on-demand resources within the deadline rather than rejecting the lease. The work studies the insight of existing backfilling algorithm that is used to schedule deadline sensitive lease in the Open Nebula cloud platform. In backfilling mechanism a lease is selected in First Come First Serve (FCFS) to be backfilled in which some ideal resources can be found out and allocated to other leases. Mainly, the scheduling performance of backfilling algorithm could be better when there are conflicts among the similar leases to backfill. Moreover, backfilling algorithm doesn't allow to schedule a new deadline sensitive task during execution. The proposed work addresses such types of issues and improves the scheduling performances with respect to maximize the task acceptance ratio and reduce the task rejection ratio. Finally, the performances are compared with the existing mechanisms that are quite good.

Offer Based Auction Mechanism for Virtual Machine Allocation in Cloud Environment

Advances in intelligent systems and computing, 2020

Nowadays, the cloud service providers provide different resources for computation such as CPU pow... more Nowadays, the cloud service providers provide different resources for computation such as CPU power, memory to a wide range of users by virtual machines (VM). The VMs vary according to their capacity of resources, where each has a limited number of resources with a different price. For this, it is very much required to allocate the available resources efficiently among the requested users, So that there should be the minimization of the cost of resources and maximization of profit from the resources. Therefore, resource allocation is one of the most challenging issues in cloud computing and fog computing. Presently, in cloud computing VMs are allocated with a fixed price mechanism which is not minimizing the cost. In this paper, to solve such problems, we purposed a hybrid combinatorial auction-based approach for allocation of VMs. Here, we calculate the cost with maximum resource utilization which is beneficiary to users on demand.

Task Scheduling Mechanism Using Multi-criteria Decision-making Technique, MACBETH in Cloud Computing

Advances in intelligent systems and computing, 2018

Cloud computing is the new era of Internet technology which provides various utilities and comput... more Cloud computing is the new era of Internet technology which provides various utilities and computing resources from the pool of resources on the basis of “pay per Use”. It is challenging one to allocate required on-demand resources for all the users’ request. Meanwhile, the service provider aims toward a better resource utilization. These user requests are called task, if task execution is bounded by time limit which is called deadline-based task. The deadline-based tasks have different parameters. To schedule these tasks, researchers proposed many works based upon these parameters. However, in this work, we considered the scheduling of deadline-based task that is a Multi-criteria Decision-making problem due to different task’s parameters associated with it. The work is proposed to implement Measuring Attractiveness through a Category-Based Evaluation Technique (MACBETH) to ranking the deadline-based task by which many tasks can meet their deadline. The results of the proposed work are quite good as compared to the existing mechanisms.

Deadline sensitive lease scheduling in cloud computing environment using AHP

Journal of King Saud University - Computer and Information Sciences, Apr 1, 2018

The OpenNebula is an open source environment which provides cloud resources with the help of Haiz... more The OpenNebula is an open source environment which provides cloud resources with the help of Haizea as a lease manager. The Haizea supports different types of leases from which deadline sensitive lease is one of them. In real time, most of the leases are deadline sensitive leases. These deadline sensitive leases are scheduled by using the backfilling algorithm. In the backfilling algorithm one of the lease is selected from the best effort queue which will provide the free resources to schedule the deadline sensitive lease. But in some scenario backfilling algorithm does not provide better scheduling if there is similar types of leases and must be in conjugative in sequence. This work aims to use AHP (Analytic Hierarchy Process) as a decision maker in the backfilling algorithm to choose the possible best lease from the given best effort queue in order to schedule the deadline sensitive lease. The proposed work improves the performance of the backfilling algorithm by scheduling more number of leases and minimizing the lease rejection using AHP.

An algorithm that earning users' trust on cloud

Now cloud computing provides many opportunities for enterprises by offering a range of computing ... more Now cloud computing provides many opportunities for enterprises by offering a range of computing services. Cloud computing has opened up a new frontier of challenges by introducing a different type of trust scenario. Today, the problem of trusting cloud computing is a paramount concern for most enterprises. It's not that the enterprises don't trust the cloud providers' intentions rather, they question cloud computing's capabilities regarding the security. To trust on the cloud, the cloud computing should have the appropriate technologies to tackle the accompanying challenges of trust. In this paper we proposed an algorithm which can increase the trust of the client on the cloud service provider. Different researchers have proposed different algorithms like data encryption and data decryption using private or public key concepts. But trust to cloud is basically based on cloud trust policies, which is not enough for the client to use cloud for the sensitive data. In this paper the proposed algorithm uses a function called key_function and the key which increases the trust to cloud during transmission. The cloud can ensure that the cloud server is providing necessary security for the user's data during transmission. By matching the key which is updated by the key_function, the user can be secured that the data accessed from the cloud server is secured.

An Improved Task Scheduling Mechanism Using Multi-Criteria Decision Making in Cloud Computing

International Journal of Information Technology and Web Engineering, Apr 1, 2019

Inthiswork,theauthorsproposeMulti-criteriaDecision-makingtoscheduledeadlinebasedtasksin cloudcomp... more Inthiswork,theauthorsproposeMulti-criteriaDecision-makingtoscheduledeadlinebasedtasksin cloudcomputing.Theexistingbackfillingtaskschedulingalgorithmcouldnothandlesimilartasks forscheduling.Inbackfillingalgorithm,tasksarebackfilledtoprovideidealresourcestoschedule otherdeadlinesensitivetasks.However,thetasktobebackfilledisselectedonfirstcome,firstserve (FCFS)basisfromschedulingqueue.Theschedulingperformancesrequiretobeimprovedwhen, therearesimilartasks.Inthisproposedwork,theauthorsproposetoimplementMCDMtechnique, TOPSIS(TechniqueforOrderPreferencebySimilaritytoIdealSolution)toimprovetheperformance of the backfilling algorithm through scheduling deadline sensitive tasks in cloud computing. It resolvestheconflictsamongthesimilartasksthatisusedasadecisionsupportsystem.Theworkis simulatedwithsyntheticdatasetsbasedonslackvaluesofthetasks.Theperformanceresultsaffirm thetaskcompletionandreductionintaskrejectioncomparedtotheexistingbackfillingalgorithm.

Truthful Resource Allocation Detection Mechanism for Cloud Computing

Resource allocation is a n-p hard problem. In spite of different proposed algorithms and methods,... more Resource allocation is a n-p hard problem. In spite of different proposed algorithms and methods, still it is a challenging one. VMs are basically considered as a resource in cloud computing. In his paper, we are considering the resource allocation in Haizea model which is an open-source VM-based lease management architecture and acts as a resource manager. Different scheduling mechanisms are used in Haizea like Advance Reservation (AR), Best effort (BE), Immediate (IM) and Deadline Sensitive (DS). Except these scheduling swapping and backfilling algorithm is proposed. But Backfilling is not applicable for all types of situation. In this work we tried to make a truthful scheduling by our proposed mechanism. The proposed mechanism also identifies where the lease can be scheduled or not. The mechanism for allocation detection is simple and robust.

A new method for weighted ensemble clustering and coupled ensemble selection

Connection science, Jan 7, 2021

ABSTRACT Clustering ensemble, also referred to as consensus clustering, has emerged as a method o... more ABSTRACT Clustering ensemble, also referred to as consensus clustering, has emerged as a method of combining an ensemble of different clusterings to derive a final clustering that is of better quality and robust than any single clustering in the ensemble. Normally clustering ensemble algorithms in the literature combine all the clusterings without learning the ensemble. But by learning the ensemble, one can define the merit of a clustering or even a cluster in it, in forming a quality consensus. In this work, we propose a cluster-level surprisal measure to define the merit of a clustering that reflects both levels of agreement as well as disagreement among clusters. Using the proposed measure of merit, we devise a polynomial heuristics that judiciously selects a subset of clusterings from the ensemble that contribute positively in forming the consensus. We also empirically show that consensus achieved by our proposed method performs better in terms of quality compared to well-known clustering ensemble algorithms on different benchmark datasets.

Modeling of Task Scheduling Algorithm Using Petri-Net in Cloud Computing

Advances in intelligent systems and computing, 2018

Task scheduling is an NP-hard problem. In earlier, different task scheduling algorithms are propo... more Task scheduling is an NP-hard problem. In earlier, different task scheduling algorithms are proposed for cloud computing environment. By adapting new and modified task scheduling algorithms, better resource utilization can be obtained. In real time, most of the tasks are deadline-based tasks. The deadline-based task has different parameters. The backfilling algorithm is used to schedule these types of tasks in Haizea. In this paper, we modeled the existing backfilling algorithm for scheduling deadline-based task using Petri-Net. The paper presents the design model of the existing backfilling algorithm. The model specifies real-time challenges of backfilling algorithm using Petri-Net. The work also comes forward with some design issues of backfilling algorithm using Petri-Net.

Petri Net: Design and Analysis of Parallel Task Scheduling Algorithm

Lecture notes in electrical engineering, Oct 29, 2017

In real time, most of the tasks are deadline based. The deadline-based task has different paramet... more In real time, most of the tasks are deadline based. The deadline-based task has different parameters: arrival time, start time, execution time, and deadline. Many performance-based task scheduling algorithms are proposed by number of researchers theoretically. But due to change of implement environment, the performance varies. Petri net is a graphical and mathematical model to evaluate and analysis of the system. In Petri net, conflicts are occurred during firing. In this paper, we designed and modeled the Petri net for scheduling deadline-based task by resolving the conflicts. We also proposed a scheduling mechanism and firing rules to schedule deadline-based tasks. The designed model increases the resource utilization of a physical system in cloud computing. The performance of the proposed model is analyzed using the PIPE v4.3.0. We analyzed the reach ability graph, convertibility graph, and steady-state analysis of the model.

Energy efficient task allocation and consolidation in multicast cloud network

Wireless Networks

Renewable energy source based quality of service (QoS)-aware routing mechanism in cloud network

Wireless Networks

JOB-DCA: A Cost Minimizing Jaya Optimization-Based Data Center Allocation Policy for IaaS Cloud Model

Advances in Intelligent Systems and Computing, 2021