21
TASK SCHEDULING ALGORITHM- A SURVEY (LOAD BALANCING IN CLOUD COMPUTING) Guided By: Prepared By: Miss Shilpi Saxena Swati Singh Assistant Professor M.Tech 2 nd Year (CSE) (Deptt. Of CS/IT) 1

Task scheduling Survey in Cloud Computing

Embed Size (px)

Citation preview

Page 1: Task scheduling Survey in Cloud Computing

1

TASK SCHEDULING ALGORITHM- A SURVEY

(LOAD BALANCING IN CLOUD COMPUTING)

Guided By: Prepared By: Miss Shilpi Saxena Swati SinghAssistant Professor M.Tech 2nd Year (CSE) (Deptt. Of CS/IT)

Page 2: Task scheduling Survey in Cloud Computing

2

CONTENTS What is Cloud Introduction to Cloud Computing Introduction to Load Balancing Literature Survey Terminology in Existing Algorithms Problem Statement References

Page 3: Task scheduling Survey in Cloud Computing

3

WHAT IS CLOUD ?“The Cloud” is simply the trendy term for a network or remote servers that can be accessed via an Internet connection store and manage information.

Deployment Model: Private Cloud Public Cloud Hybrid Cloud

Service Model: Software as a Service (SaaS) Platform as a Service (PaaS) Infrastructure as a Service (IaaS)

Page 4: Task scheduling Survey in Cloud Computing

4

DEPLOYMENT MODELPrivate Cloud: Data center architecture owned by a single company. Eg: IBM’s

Blue Cloud, Sun Cloud, Window Azure.Public Cloud: It is basically the internet. Service provider use internet to make

resource available to general people.Eg: Gmail, Office 365, Dropbox.

Hybrid Cloud: For instance during peak periods individual applications, or portion of applications can be migrate to the public cloud.

Page 5: Task scheduling Survey in Cloud Computing

5

SERVICE MODEL Software as a Service (SaaS): Email, Online Games, Facebook. Platform as a Service (PaaS): Google Compute Engine, Apache Stratos Infrastructure as a Service (IaaS): Virtual Machine, Storage, Servers

Page 6: Task scheduling Survey in Cloud Computing

6

INTRODUCTION TO CLOUD COMPUTING

Cloud computing is Internet-based computing, whereby shared resource, software, and information are provided to computers and other devices on demand.

Page 7: Task scheduling Survey in Cloud Computing

7

LOAD BALANCING

It is a process of assigning the total load over each node of the collective system to make resource utilization effective and to improve the response time of the job.Goals of load balancing: 1- to improve the performance substantially 2- to have a backup plan in case the system fail even

partially 3- to maintain the system stability

   

 T1

 T2

 T3

 T4

R1 R2

 

 T1 T3

 T2

 T4

R1 R2

Page 8: Task scheduling Survey in Cloud Computing

8

LITERATURE SURVEY Traditional Min-Min Heuristic Algorithm[1] Based on the concept of Minimum Completion Time. It works in two phase:

Expected Completion time is calculated for each task. Task with overall minimum completion time is scheduled.

Result: No longer waiting of processor for smaller tasks.

Problem: Starvation (For larger tasks).Unable to balance load when small task scheduled

initially.

Page 9: Task scheduling Survey in Cloud Computing

9

LITERATURE SURVEY Traditional Max-Min Heuristic Algorithm[1][2] Based on the concept of Maximum Completion Time. It works in two phase:

Expected Completion time is calculated for each task. Task with overall minimum completion time is scheduled.

Result: No longer waiting of processor for larger tasks.Good load balancing when smaller task are more in

number then larger task.Problem: Starvation (For smaller tasks).

Unable to balance load when number of larger task increases.

Page 10: Task scheduling Survey in Cloud Computing

10

LITERATURE SURVEY Resource Awareness Scheduling Algorithm[1][3] Hybrid algorithm (Max-Min & Min-Min) known as RASA. Based on the concept of Completion Time of each task. It works in two phase:

Expected Completion time is calculated for each task. Max-min and Min-Min is applied alternatively to schedule task. If number of resources are odd, apply Min-Min strategy first

otherwise Max-Min strategy.Result: No longer waiting of processor for larger or smaller tasks.

Good load balancing.Problem: Unable to balance load when number of larger task increases.

Page 11: Task scheduling Survey in Cloud Computing

11

LITERATURE SURVEY Qos Guided Min-Min Heuristic Algorithm[4] It adds a constraint of QoS (Quantitative QoS for

measuring bandwidth). It works in two phase:

Task are classified under high and low bandwidth. Task required high bandwidth are scheduled first.

Result: Task under lower and higher bandwidth are scheduled parallel.

Good load balancing.Problem: Unable balance load when all task satisfy low or high bandwidth.

Page 12: Task scheduling Survey in Cloud Computing

12

LITERATURE SURVEY Reliable RASA Scheduling Algorithm[7]

It added a constraint of Resource Reliability in existing Max-Min, Min-Min & RASA Algorithm (Qualitative Qos for improving makespan and load balance).

It works in three phase: Resource Reliability is provided in percentage. Resource manager sets a reliability criteria. An appropriate resource which fulfil the criteria is selected.

Result: Execution time decreases. Risk associated with job failure reduced.

Problem: Increase in reliability percentage, increase in execution time.

Increment in percentage will make only few resource available.

Page 13: Task scheduling Survey in Cloud Computing

13

LITERATURE SURVEY Improved Max-Min Task Scheduling Algorithm[1]

[5] Based on the concept of Minimum Completion Time. It works in two phase:

Task with maximum execution time is selected (Largest Task)

Selected task is scheduled over resource with minimum completion time (Slowest Resource).

Result: Makespan reduced Good load balancing.

Problem: Task with largest execution time (greater then the sum of smaller task) results in the increment of makespan.

Page 14: Task scheduling Survey in Cloud Computing

14

LITERATURE SURVEY Enhanced Max-min Task Scheduling Algorithm[6] Based on the concept of Minimum Completion Time. It works in two phase:

Expected Completion time is calculated for each task. Average or nearest greater then average task is selected. Selected task is scheduled over resource with minimum

completion time (Slowest Resource).Result: Makespan reduced.

Better load balancing.Problem: Unable to balance load when number of larger task

increases.

Page 15: Task scheduling Survey in Cloud Computing

15

PROBLEM STATEMENT

An optimized Task Scheduling Algorithm is need to design which makes:

Proper resource utilization. Reduces makespan. Reduces overhead from resource. Balance proper load over resource. Increase throughput and system performance. Better CPU utilization. Minimize turnaround, waiting and response time.

Page 16: Task scheduling Survey in Cloud Computing

16

METHODOLOGY To be proposed …..

Page 17: Task scheduling Survey in Cloud Computing

17

TOOL USED FOR SIMULATIONCloudSim: It is a simulation tool such as SimJava &

GridSim. It is used to calculate performance of the

environment. It is the only simulator which support virtualization

of Cloud Computing. To guarantee Qos in CloudSim, we need to propose

an optimized Task Scheduling Algorithm.

NetBeans: It is a Software Development platform written in Java. It allows applications to be developed from a set of

modular software components called modules.

Page 18: Task scheduling Survey in Cloud Computing

18

CONCLUSION To be proposed …..

Page 19: Task scheduling Survey in Cloud Computing

19

REFERENCES[1] S. Devipriya & C. Ramesh, "Improved Max-Min Heuristic Model for Task Scheduling

in Cloud," International Conference on Green Computing, Communication and Conservation of Energy (ICGCE), 978-1-4673-6126-2/13 2013 IEEE.

[2] Etminani .K, and Naghibzadeh. M, "A Min-min Max-min Selective Algorithm for Grid Task Scheduling,“ Third IEEE/IFIP International Conference on Internet, Uzbekistan, 2007

[3] SaeedParsa and Reza Entezari-Maleki, "RASA: A New Grid Task Scheduling Algorithm," International Journal of Digital Content Technology and its Applications,Vol. 3, pp. 91-99, 2009.

[4] He. X, X-He Sun, and Laszewski. G.V, "QoS Guided Min-min Heuristic for Grid Task Scheduling," Journal of Computer Science and Technology, Vol. 18, pp. 442-451, 2003

[5] O. M. Elzeki, M. Z. Reshad, M. A. Elsoud, "Improved Max-Min Algorithm in Cloud Computing," International Journal of Computer Applications (0975 – 8887), Volume 50 – No.12, July 2012.

[6] Upendra Bhoi1, Purvi N. Ramanuj, “Enhanced Max-min Task Scheduling Algorithm in Cloud Computing,” International Journal of Application or Innovation in Engineering & Management on Volume 2, 2013.

Page 20: Task scheduling Survey in Cloud Computing

20

REFERENCES[7] Sunilkumar Nakum, C. Ramakrishna, Amit Lathigara, “Reliable RASA Scheduling

Algorithm for Grid Environment,” 2014 IEEE International Conference on Computer Communication and Systems(ICCCS '14), Feb 20-21, 2014.

[8] Sung-Min Jung, Nam-Uk Kim, Tai-Myoung Chung, “Applying Scheduling Algorithms with QoS in the Cloud Computing,” 978-1-4799-0604-8/13 2013 IEEE.

[9] Yash P. Dave, Avani S. Shelat, Dhara S. Patel, Rutvij H. Jhaveri, “Various Job Scheduling Algorithms in Cloud Computing: A Survey,” ICICES2014 - S. A. Engineering College, Chennai, ISBN No.978-1-4799-3834-6/14 2014 IEEE

Page 21: Task scheduling Survey in Cloud Computing

21

THANK YOU