Improved Community Aware Job Scheduling Algorithm in Cloud

Improved Community Aware Job Scheduling Algorithm in Cloud

1. Vivekanandan, Student Master of computer applications B.S.Abdur Rahman University

   Chennai, India

   vivekprahalathan@gmail.com

   Mrs.S.Prathiba,Assitant Professor Department of Computer Applications , B.S.Abdur Rahman University

   Chennai, India

   prathibarajiv@gmail.com

   AbstractA variety of computing concepts that involve a large number of computers connected through a Real-time communication network such as the Internet. . In this paper, an improved community aware scheduling algorithm is proposed in which the resources are utilized effectively by minimizing the execution time. A task scheduling model is established for reducing the system power consumption of cloud computing and to improve the profit of service providers

   Keywords: Cloud computing, resource

   1. INTRODUCTION

      Cloud computing, to put it simply, means Internet Computing. The Internet is commonly visualized as clouds; hence the term cloud computing for computation done through the Internet. With Cloud Computing users can access database resources via the Internet from anywhere, for as long as they need, without worrying about any maintenance or management of actual resources. Besides, databases in cloud are very dynamic and scalable. The best example of cloud computing is Google Apps where any application can be accessed using a browser and it can be deployed on thousands of computer through the Internet Cloud computing provides the facility to access shared resources and common infrastructure, offering services on demand over the network to perform operations that meet changing business needs. The location of physical resources and devices being accessed are typically not known to the end user. It also provides facilities for users to develop, deploy and manage their applications on the cloud, which entails virtualization of resources that maintains and manages itself. As a matter of fact, data is the most important part of cloud computing; thus, data security is the top most priority in all the data operations of cloud. Here, all the data are backed up at multiple locations. This astoundingly increases the data storage to multiple times in cloud compared with a regular system. Redundancy of data is crucial, which is a must-have attribute of cloud computing. In this paper a job scheduling algorithm is proposed in which the status of the job is known to the user periodically so the user can reschedule the job in case any failure occurs. The paper is organized as follows. In Section

      2 related work is discussed, section 3 discuss about

      proposed model, section 4 discuss about results, conclusion is discussed in section 5.

   2. RELATED WORK

      In paper (1) a community aware scheduling algorithm developed but before execution the jobs are assigned to the suitable machine and that machine will continue executing the jobs with interpretation. In job scheduling main focus is to find approite reasons of the job that is used schedule the job. In paper(1) a community aware scheduling job scheduling proposed in which the user is unaware of status of the jobs cannot be rescheduled due to unique administrative management and system failure. In this algorithm, jobs are executed according to the order of job arriving time. The next job will be executed in turn. The FCFS algorithm [20] may induce a convoy effect. The convoy effect happens when there is a job with a large amount of workload in the job queue. When this occurs, all the jobs queued behind it must wait a long time for the long job to finish. The RR algorithm [21] mainly focuses on the fairness problem The RR algorithm defines a ring as its queue and also defines a fixed time quantum. Each job can be executed only within this quantum, and in turn. If the job cannot be completed in one quantum, it will return to the queue and wait for the next round. The major advantage of RR algorithm is that jobs are executed in turn and do not need to wait for the previous job completion. Therefore, it does not suffer from a starvation problem. However, if the job queue is fully loaded or workload is heavy, it will take a lot of time to complete all the jobs. Furthermore, a suitable time quantum is difficult to decide.Job scheduling is most important task in cloud computing environment because user have to pay for resources used based upon time. Job scheduling has didnt overcome issues such as bottleneck, single point failure, and impractical unique administrative management. In Section [2] discuss about related work it led to the failure of total system. Its disadvantage is No Standard scheduling techniques is used, Deadlock and failure of job occurs, No rescheduling process is done Any

      job gets failed the total system gets collapsed,User doesnt

      know job status.

   3. PROPOSED MODEL

      In the proposed system an improved community aware scheduling algorithm (ICSA) is proposed in which the resources are utilized very productively. The proposed scheduling algorithm allows all the users to access the server anytime.The advantages of the proposed system are If any job is delayed, rescheduling process is done on time. The time interval of the job is known to the user.The algorithm for job rescheduling is represented . Once a node is selected as the assignee node for delegated jobs execution, numerous events such as newly arrived jobs and resources are able to keep changing the status of the overall cloud dynamically through time. Therefore, if the assigned job is still not executed yet after some time, the current assignee node may not remain an optimal choice for the assigned jobs execution because some other nodes could offer a shorter job response time because of their volatile resource usages. In this case, each node of the cloud needs to check periodically whether some of the already queued jobs could be re-assigned to other nodes for obtaining better performance in terms of job execution and cloud resource utilization.

      1. Job Rescheduling

         Once a node is selected as the assignee node for a delegated

         jobs execution, numerous events such as newly arrived jobs

         and resources are able to keep changing the status of the overall cloud dynamically through time. Therefore, if the assigned job is still not executed yet after some time, the current assignee node may not remain an optimal choice for the assigned jobs execution because some other nodes could offer a shorter job response time because of their volatile resource usages.

         B.JobQueue

         A job queue is a data structure maintained by job scheduler software containing jobs to run. Users submit their programs that they want executed, "jobs", to the queue for batch processing.

         C.The architecture of cloud computing

         Cloud computing is Internet-based development and use of computer technology. The cloud is a metaphor for the Internet (based on how it is depicted in computer network diagrams) and is an abstraction for the complex Infrastructure it conceals. The architecture behind cloud computing is a massive network of "cloud resources" interconnected as if in a grid running in parallel, always using the technique of virtualization to maximize the computing power per server Applications of users work on the virtual operation systems.

         FIGURE1: ARCHITECTURE DIAGRAM OF THE PROPOSED SYSTEM

         TABLE1:THE NOTATION USED IN ALOGRITHM THE ALGORITHM IS AS FOLLOWS

         Symbols	Description
         n	Requester node
         <>j,	Which needs to be allocate
         req,	Estimation message generated for job
         xn	Length of the known node.an arbitrary node from list
         Ecpue1 Ememe2 Ediske3 Enwe4	Estimation cpu Estimation memory Estimation disk Estimation network
         Ex	Ex taking the values 0 Ex 100 and e1, e2, e3, e4 are basis vectors.
         Estimation time	User for confirmation message to proceed job or not

         ALGORITHM

         Step 1:

         Request is submitted.

         Step 2:

         Authentication is done for each request

         Step 3:

         Authenticated request is placed in job queue

         Step 4:

         Estimation time is calculated based on job traffic and confirmation message is sent to the user.

         Step 5:

         Job is executed successfully and periodic update is given every 15 minutes

         .

         Step 6:

         If job is not executed successfully, the request is rescheduled .

         FIGURE 2: PROPOSED ALOGRTHIM

   4. RESULTS

      The simulation is based on the cloud simulation tool kit.. The heterogeneous environment is build by various resource specification according to the cloud computing..The resources in the cloud differ by operating system, cpu speed, memory and bandwidth.

      FIGURE3: LOGIN PAGE

      FIGURE4: FIELDS

      FIGURE5: JOB SUBMISSION

      FIGURE6: DISPLAYING JOB

      FIGURE7:VIEW JOB

      FIGURE8:JOB ASSIGN

      FIGURE9:CALCULATING JOB

      FIGURE10: JOB QUEUE

      FIGURE11:NODE CREATION

   5. CONCLUSION

      In this paper an improved community aware job scheduling algorithm is proposed in which the resources are scheduled very effective. Mainly we have to reduce traffic. Finally we got good job scheduling.

   6. ACKNOWLEDGMENT

I would like to thank my project supervisor Mrs. S.Prathiba, without her help and guidance this project would not have completed and would also like to HOD, COMPUTER APPLICATIONS Department and other Professors for extending their help & support in giving technical ideas about the paper and motivating to complete the project successfully.

REFERENCES

1. Ye Huang, Nik Bessis, Exploring decentralized dynamic scheduling for clouds and clouds using the Community-aware scheduling algorithmJournal of Computer and System

   Sciences Y. Huang et al. / Future Generation Computer Systems 29 (2013) 402415

2. Yun-Han Lee,, Seiven Leu,, Ruay-Shiung Chang, Improving job scheduling algorithms in a cloud environment, Y.-H. Lee et al. / Future Generation Computer Systems 27 (2011) 991998

3. I. Forster, C. Kesselman, S. Tuecke, The anatomy of the cloud: enabling scalable

   virtual organization, International Journal of Supercomputer Application 15 (2001) 200222.

4. R. Bolze, F. Cappello, E. Caron, M. Daydé, F. Desprez, E. Jeannot, Y. JégouS. Lanteri, J. Leduc, N. Melab, et al., Cloud5000: a large scale and

   highlyreconfigurable experimental Cloud testbed, International Journal of High

   Performance Computing Applications 20 (4) (2006) 481.

5. C. Catlett, P. Beckman, D. Skow, I. Foster, Creating and operating national-scale

   cyberinfrastructure services, CTWatch Quarterly 2 (2) (2006) 210.

6. W. Gentzsch, D-cloud, an e-science framework for German scientists, in:ISPDC06: Proceedings of the Proceedings of The Fifth International Symposiumon Parallel and Distributed Computing, IEEE Computer Society, Washington,

   DC, USA, 2006, pp. 1213. doi:10.1109/ISPDC.2006.16.

7. F. Gagliardi, The EGEE European cloud infrastructure project, High PerformanceComputing for Computational Science-VECPAR 2004, 2005, pp. 194203.

8. B. Chun, D. Culler, T. Roscoe, A. Bavier, L. Peterson, M. Wawrzoniak,M. Bowman, Planetlab: an overlay testbed for broad-coverage services, ACMSIGCOMM Computer Communication Review 33 (3) (2003) 12.

9. D. Feitelson, A. Weil, Utilization and predictability in scheduling the IBMSP2 with backfilling, in: 12th International Parallel ProcessingSymposium,Citeseer, 1998, pp. 542546.

10. D. Feitelson, Packing schemes for gang scheduling, in: Job Scheduling Strategies for Parallel Processing, Springer, 1996, pp. 89110.