Presentation By
Dr. Manjunath Kotari
Professor & Head-CSE
AIET, Moodbidre
Scheduling in Cloud Computing
01/20/191
Contents
Introduction
Traditional Cloud Scheduling
Job Scheduling Framework in Clouds
Dynamic Cloud Scheduling
Fault-Tolerant Scheduling in Cloud
Deadline-Constrained Dynamic Scheduling.
Inter-Cloud Meta-Scheduling
Conclusion
References
01/20/192
Introduction
Cloud computing process the huge data.
Scheduling mechanism is essential.
Scheduling is at the heart of Distributed Computing.
PaaS model -> Workflow (job) Scheduling
IaaS model -> Virtual Machines (VM) Scheduling.
Scheduling decides allocation of VMs.
An effective scheduling can
Reduce operational costs
Reduce queue waiting time
Increase resource utilization
01/20/193
Traditional Cloud Scheduling
FCFS
Simplest
Lower Headache
Fit for Small Size of data
No Starvation
Priority Queue Scheduling
Avoid starvation
Round Robin Scheduling
Time quantum is used
01/20/194
Traditional Cloud Scheduling [cont..]
Shortest Job First Scheduling
Basis of shortest execution time
Multi Level Feedback Queue Scheduling
Use multiple queue with RR & FCFS.
Multi Level Queue Scheduling
Uses multiple queues with different scheduling.
01/20/195
Job Scheduling Framework in Clouds
Challenges
Allocating massive job requests
Satisfying user QoS requirements.
Maintaining average response time
01/20/196
Job Scheduling Framework
01/20/197
Steps
User portal -manages job requests.
Job scheduler - routing decisions & selects VM instance.
Management module-
VM Monitor
Job monitor - keeps track of jobs
Job profiling - identify job types
History repository - stores the records of job .
01/20/198
Dynamic Cloud Scheduling
Classification based on Historical data.
Creation of VMs
Matches tasks with suitable VMs dynamically
Minimize task waiting time and executing time.
01/20/199
Cloud task scheduling framework
01/20/1910
Fault-Tolerant Scheduling in Cloud
Fault tolerance technique
Enhances Reliability and Availability
Introduce redundancy
Incurs extra overhead.
01/20/1911
Fault-tolerant scheduling architecture
01/20/1912
Steps
Global Scheduler
 Analyzes information
Makes decisions
Sends the primary/backup copies of the task to different VMs.
Local Scheduler
Rearranging the order of the local queue
Resource Manager
Decides how VMs should be added or migrated
01/20/1913
Deadline-Constrained Dynamic Scheduling
01/20/1914
Inter-Cloud Meta-Scheduling
Multiple autonomous clouds.
Functions under a single federated management entity.
The algorithm estimates the queue length of neighboring
processors
Reschedules the loads based on estimates.
The method aims to increase the possibilities to gain load
balancing.
01/20/1915
Inter-Cloud Meta-Scheduling
Facilitates scalable resource provisioning.
ICMS is based on a novel message exchange
mechanism.
Offers improved flexibility, robustness and
decentralization.
01/20/1916
Conclusions
Scheduling and execution improve service quality of the
clouds.
Creates VMs and decrease the failure rate of task
scheduling.
Increase in resource utilizations
01/20/1917
References
1. PeiYun Zhang, and MengChu Zhou, Dynamic Cloud Task Scheduling Based on
a Two-Stage Strategy, IEEE TRANSACTIONS ON AUTOMATION
SCIENCE AND ENGINEERING, VOL. 15, NO. 2, APRIL 2018.
2. YI WEI , LI PAN, SHIJUN LIU , LEI WU, AND XIANGXU MENG, DRL-
Scheduling: An Intelligent QoS-Aware Job Scheduling Framework for
Applications in Clouds, IEEE ACCESS, 2018.
3. PENGZE GUO, MING LIU1, JUN WU, ZHI XUE, AND XIANGJIAN HE,
Energy-Efficient Fault-Tolerant Scheduling Algorithm for Real-Time Tasks
in Cloud-Based 5G Networks, IEEE ACCESS ,2018.
4. Jyoti Sahni and Deo Prakash Vidyarthi, A Cost-Effective Deadline-
Constrained Dynamic Scheduling Algorithm for Scientific Workflows in a
Cloud Environment, IEEE TRANSACTIONS ON CLOUD COMPUTING,
VOL. 6, NO. 1, JANUARY-MARCH 2018.
01/20/1918
References [cont..]
5. Stelios Sotiriadis , Nik Bessis, Ashiq Anjum, and Rajkumar Buyya, An Inter-
Cloud Meta-Scheduling (ICMS) Simulation Framework: Architecture and
Evaluation, IEEE TRANSACTIONS ON SERVICES COMPUTING, VOL.
11, NO. 1, JANUARY/FEBRUARY 2018.
6. MIAN GUO, QUANSHENG GUAN, AND WENDE KE, Optimal Scheduling
of VMs in Queuing Cloud Computing Systems With a Heterogeneous
Workload, IEEE ACCESS 2018.
7. Ruiting Zhou , Zongpeng Li, and Chuan Wu,Scheduling Frameworks for Cloud
Container Services, IEEE/ACM TRANSACTIONS ON NETWORKING,
VOL. 26, NO. 1, FEBRUARY 2018
8. Arnav Wadhonkar , Deepti Theng ,A Survey on Different Scheduling
Algorithms in Cloud Computing, International Conference on Advances in
Electrical, Electronics, Information, Communication and Bio-Informatics
(AEEICB16).
01/20/1919

Scheduling in cloud

  • 1.
    Presentation By Dr. ManjunathKotari Professor & Head-CSE AIET, Moodbidre Scheduling in Cloud Computing 01/20/191
  • 2.
    Contents Introduction Traditional Cloud Scheduling JobScheduling Framework in Clouds Dynamic Cloud Scheduling Fault-Tolerant Scheduling in Cloud Deadline-Constrained Dynamic Scheduling. Inter-Cloud Meta-Scheduling Conclusion References 01/20/192
  • 3.
    Introduction Cloud computing process thehuge data. Scheduling mechanism is essential. Scheduling is at the heart of Distributed Computing. PaaS model -> Workflow (job) Scheduling IaaS model -> Virtual Machines (VM) Scheduling. Scheduling decides allocation of VMs. An effective scheduling can Reduce operational costs Reduce queue waiting time Increase resource utilization 01/20/193
  • 4.
    Traditional Cloud Scheduling FCFS Simplest LowerHeadache Fit for Small Size of data No Starvation Priority Queue Scheduling Avoid starvation Round Robin Scheduling Time quantum is used 01/20/194
  • 5.
    Traditional Cloud Scheduling[cont..] Shortest Job First Scheduling Basis of shortest execution time Multi Level Feedback Queue Scheduling Use multiple queue with RR & FCFS. Multi Level Queue Scheduling Uses multiple queues with different scheduling. 01/20/195
  • 6.
    Job Scheduling Frameworkin Clouds Challenges Allocating massive job requests Satisfying user QoS requirements. Maintaining average response time 01/20/196
  • 7.
  • 8.
    Steps User portal -managesjob requests. Job scheduler - routing decisions & selects VM instance. Management module- VM Monitor Job monitor - keeps track of jobs Job profiling - identify job types History repository - stores the records of job . 01/20/198
  • 9.
    Dynamic Cloud Scheduling Classificationbased on Historical data. Creation of VMs Matches tasks with suitable VMs dynamically Minimize task waiting time and executing time. 01/20/199
  • 10.
    Cloud task schedulingframework 01/20/1910
  • 11.
    Fault-Tolerant Scheduling inCloud Fault tolerance technique Enhances Reliability and Availability Introduce redundancy Incurs extra overhead. 01/20/1911
  • 12.
  • 13.
    Steps Global Scheduler  Analyzesinformation Makes decisions Sends the primary/backup copies of the task to different VMs. Local Scheduler Rearranging the order of the local queue Resource Manager Decides how VMs should be added or migrated 01/20/1913
  • 14.
  • 15.
    Inter-Cloud Meta-Scheduling Multiple autonomousclouds. Functions under a single federated management entity. The algorithm estimates the queue length of neighboring processors Reschedules the loads based on estimates. The method aims to increase the possibilities to gain load balancing. 01/20/1915
  • 16.
    Inter-Cloud Meta-Scheduling Facilitates scalableresource provisioning. ICMS is based on a novel message exchange mechanism. Offers improved flexibility, robustness and decentralization. 01/20/1916
  • 17.
    Conclusions Scheduling and executionimprove service quality of the clouds. Creates VMs and decrease the failure rate of task scheduling. Increase in resource utilizations 01/20/1917
  • 18.
    References 1. PeiYun Zhang,and MengChu Zhou, Dynamic Cloud Task Scheduling Based on a Two-Stage Strategy, IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING, VOL. 15, NO. 2, APRIL 2018. 2. YI WEI , LI PAN, SHIJUN LIU , LEI WU, AND XIANGXU MENG, DRL- Scheduling: An Intelligent QoS-Aware Job Scheduling Framework for Applications in Clouds, IEEE ACCESS, 2018. 3. PENGZE GUO, MING LIU1, JUN WU, ZHI XUE, AND XIANGJIAN HE, Energy-Efficient Fault-Tolerant Scheduling Algorithm for Real-Time Tasks in Cloud-Based 5G Networks, IEEE ACCESS ,2018. 4. Jyoti Sahni and Deo Prakash Vidyarthi, A Cost-Effective Deadline- Constrained Dynamic Scheduling Algorithm for Scientific Workflows in a Cloud Environment, IEEE TRANSACTIONS ON CLOUD COMPUTING, VOL. 6, NO. 1, JANUARY-MARCH 2018. 01/20/1918
  • 19.
    References [cont..] 5. SteliosSotiriadis , Nik Bessis, Ashiq Anjum, and Rajkumar Buyya, An Inter- Cloud Meta-Scheduling (ICMS) Simulation Framework: Architecture and Evaluation, IEEE TRANSACTIONS ON SERVICES COMPUTING, VOL. 11, NO. 1, JANUARY/FEBRUARY 2018. 6. MIAN GUO, QUANSHENG GUAN, AND WENDE KE, Optimal Scheduling of VMs in Queuing Cloud Computing Systems With a Heterogeneous Workload, IEEE ACCESS 2018. 7. Ruiting Zhou , Zongpeng Li, and Chuan Wu,Scheduling Frameworks for Cloud Container Services, IEEE/ACM TRANSACTIONS ON NETWORKING, VOL. 26, NO. 1, FEBRUARY 2018 8. Arnav Wadhonkar , Deepti Theng ,A Survey on Different Scheduling Algorithms in Cloud Computing, International Conference on Advances in Electrical, Electronics, Information, Communication and Bio-Informatics (AEEICB16). 01/20/1919