The Latest Review: Scheduling Algorithms On The Cloud Computing Based Environment

The Latest Review: Scheduling Algorithms On

The Cloud Computing Based Environment

Ashish, Prof. M.M.S. Rauthan, Varun Barthwal, Rohan Varma

Abstract: The cloud computing framework incorporates a few parts, for instance, customers or clients, server farms named data centers a nd distributed servers. It is reasonable for the social event the computational prerequisites of an enormous estimated task. The optimal timeliness of resources or available asset allocation and tasks is just a central depending issue in a cloud computing-based environment, as the supplier who serves multiple customers in a cloud situation. Many research experts in this field have created various scheduling algorithms and used trending to overcome these issues. As the algorithm developed and investigated its performance under various parameters, another issue comes to the researcher to select the optimal determination algorithm to solve the specific problem. The major work of scheduling algorithms is to increase the use of accessible resources for energy consumption and power management etc. They then used the migration process to limit the preparation time for t he task. Through this review of research on the advancement of scheduling in cloud computing, we are trying to attempt to cover each of the previous research and their exams on their Quality of Services (QoS).

Keywords: Scheduling Algorithm, (FCFS) First Come First Serve, (RR) Round-Robin Scheduling, Max-Min Algorithm, Priority-Based Job Scheduling, Genetic Algorithm.

——————————  ——————————

1.

INTRODUCTION

Advancement in technological innovation and how current societies and organizations adapt and prop up in this cloud era are clear evidence that cloud computing is the future stage for individual and organizational needs as well. [1] Cloud computing has come to be one of the almost all appealing ground in both ICT (Information and Communication Technology) and scholarly researches. [2]The fundamental purpose for this is most associations, organizations, and individuals can't purchase such costly assets (equipment and programming software environment). Cloud computing is increasingly transforming into the best stage for associations and individual; with this figuring advancement, clients can get ready to access various facility, approach, store and network system basic registering assets, working structures, for example, working frameworks or operating system, virtual work territories, web administrations and services, databases and research and development platform at minor expenses.[3]It additionally uses explicit applications because "you pay as you go to the service" offered by the cloud computing environment. Many different supports of cloud computing environments include cost savings, adaptability, high availability, fast use, energy productivity and efficiency, and versatility. [4, 5, 6] We omit banters about our definition and use the worldwide definition announced by NIST [7], because "cloud computing is a design that facilitates suitable processing assets (e.g., network, server, Storage) is convenient for a demand system approach to a normal pool. Entities, applications, and services that can be speedily provisioned and free with minimum management attempt or service supplier interaction.

―The effect of cloud computing as technological innovations can be felt by our day to day practice. It has affected our lives in a wide range of development which has given the importance and message of how to create computer processing services. [8] Monitoring or planning of accessible resources, to limit customer consumption and advance execution on computing resources, to deliver interest services to customers.

2 LITERATURE SURVEY

Cloud computing term is a developing innovative technology or method that is becoming widespread as it is able to access computing resources, for example, online applications, storage units, computer games, services, motion pictures or movies and music on demand. In such a manner cloud computing require the management procedure or algorithm that landing as multiple queues. The appearance of the work is irregular with its various classifications and types. All of these require a scheduling algorithm to operate at a constant state with unusual reference for function control, and effectiveness. The previous years have seen an energizing development for cloud task scheduling and resource allotment. It has maintained its growth in the computer systems, media transmission, and storage capacity of markets, and has improved to its concentration with the presentation of innovative technology and well-crafted scheduling algorithms by various researchers, for example, Max-Min, Min, Min-Max, FCFS, priority-based algorithm and task scheduling allocated resources and cloud is the wrong job scheduling for fast producers. In this segment, we are exploring the past examinations by numerous researchers those have presented different scheduling algorithm from timely.

3 SCHEDULING ALGORITHMS

Scheduling alludes to the organization of standards, strategies, and mechanisms that oversee the solicitation where assets are determined for different procedures and the undertaking must be terminated. When a process hangs for input or output requests or a few separate requests required for its execution, the CPU will be seated in idle mode. The CPU is allocated to another process that is already in the ready queue to increase CPU usage time. This can only be accomplished by scheduling. Scheduling algorithms straight ————————————————

 Ashish is currently pursuing an M.Tech program in (CSE) in H.N.B.G.U. Srinagar(Garhwal) Uttarakhand (India),

E-mail: er.a.muyal@mail.com

 Prof. M.M.S. Rauthan is currently Professor & Head Dept. of CSE in H.N.B.G.U. Srinagar(Garhwal) Uttarakhand (India),

 E-mail: mmsrauthan@rediffmail.com

 Varun Barthwal is currently Assistant Professor Dept. of IT in H.N.B.G.U. Srinagar (Garhwal) Uttarakhand (India),

E-mail: varuncsed1gmail.com

 Rohan Verma is currently Assistant Professor Dept. of CSE in H.N.B.G.U. Srinagar (Garhwal) Uttarakhand (India),

2388

influence the capacity of customers' undertakings and additionally make efficient use of resources in cloud computing environments that carry out tasks. Consequently, in what way to understand the ideal distribution access of customers' tasks can be an unresolved query for tasking scheduling for a particular environment. In the field of cloud computing, the algorithm of task realization has as come behind: Initially, assets and tasks are mapped in relation to the current assignment of resources and data according to fundamental approaches or techniques. Until then, tasks are mapped between the cloud client's service quality and general resources can be distributed for the use of tasks to firstly confirm the fitness of the tasks. Keeping everything in mind, the sequence of results is executed by presenting the customer based requirement. [9, 10] In this paper, we chose five scheduling algorithms for literature that comprise of (FCFS), Max-Min, Round-Robin, Priority-based job scheduling algorithm, and Genetic Algorithm. Remain section of this paper are cover the comparison of the above-mentioned scheduling structure algorithms. In this review paper, we selected five scheduling algorithms for the literature. The remainder of this paper actually covers comparisons of the above scheduling algorithms.

3.1FIRST COME FIRST SERVER (FCFS) SCHEDULING

ALGORITHM

Random and Backfill Hamscheret. al.[11] have talked about specific heuristic-based algorithms such as the FCFS algorithm Random and Backfill. For estimating this algorithm, he has stimulated. In FCFS, tasks are defined and executed in "their submission request". Additionally, in the event that at current, the particular machines are not empty, the scheduler needs to hold on until the start of the job. In random scheduling, the resulting function for scheduling is determined arbitrarily, because of this, scheduling is a "non-deterministic". Also, into the unsystematic strategy, there is no inclination for tasks. In any case, the backfill algorithm may be additionally treated as "out-request rendering of FCFS", a plan to maintain a strategic distance from unnecessary idle time due to long-term jobs.

3.2MAX-MIN (MAXIMUM-MINIMUM)ALGORITHM

Max-Min is such a resource assignment and further scheduling algorithm used in cloud computing and grid-based computing to make pans, costs and benefits, and in the environments needed to limit resource utilization. Bhoi and Ramanuj, [12] exhibited an upgraded Max-Min tasked scheduling in cloud computing in which they considered choosing normal or closest more noteworthy than normal tasks as opposed to choosing and allotting enormous assignment first as consistently on account of the conventional Max-Min scheduling algorithm which saw a generous decrease in makes pan and furthermore adjusting load successfully when the outcomes was actualized. (Mao, et. al). [13] With (Li, et. al). [14] Set maximum-minimum calculations for the task scheduling to accommodate exact loads of a flexible cloud. The stipulate algorithm secured a work status surface for approximate the going on the outstanding burden of Virtual Machines and the uncertain implementation time of enterprises. Simulation results show that the max-min algorithm increases the use of assets and decrease the use of times for task schedule. Wang and Yu [15] propose a min-min modified high-performance scheduling algorithm for the tasks scheduled to enhance the organization of the (cloud computing framework). Be that as it may, the min-min algorithm satisfies the maximum and complete

throughput times for a continuous task and is then depicted for scheduling to essentially complete in a short time frame. The result is visible that computations are used for the job schedule for cloud computing. Zhang with Xu[16] recommend a minimum-minute tasks schedule algorithm dependent on the quality of services (QoS) imperative in cloud computing. Comparison of proposed computation resources or tasks Estimates and later tells customers to fulfill their requests. Demonstrate the results that basic-QoS-min-min outperforms the execution time and QoS fulfillment as opposed and as the minimum-min algorithm needs in cloud computing. (Tsai, et al). [17] Suggest breed breeding techniques make up the minimum-minimum or lengthy in work firstly (LJF) longest job first, to decrease the make pan to determine the task in the odd grid condition. Simulation results verify that the presenting of the suggested system is superior to the difference in decrease make pan time. (Chen, et al). [18] Makeup in cloud computing introduces two novel scheduling algorithms to improve resource utilization and customer needs. The (LBIMM ) algorithms and the (PA-LBIMM) algorithms depend on the Min-Min algorithm. Simulations result are seen that both (LBIMM) and (PALBIMM) algorithms are out in the performance, which is superior to the fundamental minimum-minute algorithm to improve consumption times, load adjustments and customer needs.

3.3ROUND ROBIN ALGORITHM

Helmy with Dekdauk [19] introduced (RR) burst round-robin, which presents relative offer scheduling algorithms in the attempts to consolidate the lower scheduling above the RR algorithm and support the concise task. (Mohanty et al). [20] The proposed (SRBRR) short resting bursts round-robin scheduling algorithms engage the processors to use dynamic time quantum to form forms with a brief outstanding burst in the round-robin way. (Yaashuwanth with Ramesh) [21] Also created another algorithm for scheduling that uses intelligent time slices for Robin planning tasks for continuously working frameworks (Mostafa et al). [22] Proposed for the discovery of a better amount (quantum) of RR CPU scheduling algorithms when all are stated in registered frameworks using number programming. (Yadav et al). [23] RR and SJF, Which proposed another calculation. From the investigation, the results demonstrate that this mixture is superior to unreserved RR. (Panda with Bhoi) [24] Suggests compelling round-robin algorithms, using the Minimum-Max scattering ratio of residual CPU (central processing unit) bursts time. This calculation outperforms the (RR) as far as normal turnaround time, normal holding time interval and a particular number of setting switches techniques. The weighted round-robin is another drawn close [25] introduction with the order of settling all recurring tasks to deactivate the VM. The weighted round-robin algorithm was conceived through the customary round robin. The proposed round robin designates functions for assets using the round-robin style, although the traditional round robin's account relies on the heaviness of payment demand rather than the current stack of virtual machines. Although constrained parameters were used in the results test, yet, the weighted round-robin was seen to be one of the best exhibitions related to time in the tested results.

3.4PRIORITY-BASED JOB SCHEDULING ALGORITHM

the drawback of the traditional weight proper scheduling algorithms. The weight just scheduling algorithms differentiates the administration of every one dynamic line based on the weight of each business stream. At the sharp point end when a new location shows the classifier attributes jobs to different categories. At that particular point, the cradle is checked for every classification and in the event that the cushion is not over-burdened, employment is kept away at that point, with the occupation usually removed. Each activity enters an alternate virtual line. Weight term, Dispatch term, Discard and ROB terms are the four principle standards of this calculation. The fundamental favorite position of this calculation is that it has introduced the ransom rule as well as leaving the standard. Exams are performed on to (NS-2) programming into the reproduce the (SRPQ-CBWFQ) algorithms. This new algorithm joins to support executives and line booking and ensures less distortion of bus continuous applications. Also, it was thought of as decency and better use of cushions. Transmission efficiency has two exceptional points of interest in this algorithm, throughput allocation and distortion without reducibility. (Chtourou, H., and Haouari, M). [27] Proposed a two-step requirement standard-based algorithm to solve the (RCPSP) resource forced task scheduling problem. These algorithms introduce a two-order algorithm for strongly resources bound enterprise scheduling. The initial stage explains the RCPSP to limit the makepan using a rule-based approximation. The second phase is expected to discover the heartiest timetable that does not exceed the limit found in the first phase. Both phases are considered as two phases. In the single Phase, every cycle consists of three stages: 1. Esteem given by the chosen requirement rule is required. 2. Unilateral choice of qualified practice according to your determination possibilities. 3. The chosen exercise is planned for the assets. In Phase II, the same emphasis is given in Phase I. Each cycle begins with the execution of further iterations that allow the task makepan to be fixed. Reverse resorption is performed to obtain the most recent achievement time of each one activity. This above advance is made only the makepan is a not sizeable than the edge resolute into step (I). Agarwal, Dr., and SaloniJain [28] his work he named CC, the generalized priority algorithm (GPA), for the wrong scheme for the new algorithm. The calculations were tested and vice versa and the FCFS and RR algorithms were used for the changing number of VMS and the remaining functions as follows and the CloudSense test system was used. The results demonstrate that propose algorithms were other productive than (RR) or (FCFS) based algorithms. Pal, A. J. [29] has created a modified prioritized deadline scheduling type of algorithm (MPDSA), which proposes to use board algorithms for productive employment execution with cutoff time limits for client businesses. MPDSA performs tasks with the nearest cutoff time postpone in a cyclic mode using dynamically times slot quantum. It expects each one activity to are characterized through its process -id, cutoff time, burst-time with arrival-time. This amount of time is retrieved by estimating the LCM of all erupted time. At that point, businesses with the least amount of distortion are chosen for execution. If the jobs are delayed at the matching time, next the (FCFS) first-come-first-served calculation is pre-owned for scheduling. Employees depend on premature quantum and if a business finishes its execution prematurely, that activity is erased from the line. This algorithm meets the requirements of the framework and supports adaptability under the heavy load of the remaining workloads.

3.5GENETIC ALGORITHM

GE Junwei [30] has demonstrated a stable genetic algorithm, considering the time to complete all the tasks, the culmination of the general undertaking, and the cost being mandatory. One of the problems in scheduling is to estimate the correct property for the show's functions. This dynamically scheduling procedure is to think about the poor every time some enterprises land together. For that reason. (S. Ravichandran with D. E. Naganathan) [31] designed a framework to stay away from this issue, enabling the shown tasks to be placed in a queue and the plan would be remodeled and these tasks would be serialized. Thus, the scheduling ends by assuming the primary assignment from the queue and an appeal is made to the property that would be best suited for GA use. The goal of this framework is to increase the utilization of assets as it additionally reduces implementation time. R. Kaur with S. Kinger [32] has introduced a task scheduling algorithm build correction (GA). They make use of another welfare potential dependent on to signify with impressive signify qualities. They guarantee those these algorithms can be available executed on both assignments with supplies planning. (Z. Zheng), et al. [33] has presented an algorithm relying on to (GA) to manage scheduling issues into a cloud computing situation called (PGA) parallel computing genetic algorithm to scientifically raise or subs-stream cloud computing scheduling issues. S. Singh [34] has given an in-depth idea about GA by introducing some changes to the task design for this cloud computing situation. They have started a calculation with care for work scheduling issues by adjusting the (GA) genetic algorithm in which the introductory populations are created to deal with rapidly achieving ideal results in the form of "popups" by way of Max-Min. (V.V. Kumar with S. Palaniswami) [35] Has presented an investigation that brings into focus on expanding the organization of the task scheduling algorithms the continuous clouds registration administration. In addition, he has introduced an algorithm using turnaround time to account for the high need for early-time work and the low need for premature birth issues.

4 COMPARISON OF SCHEDULING ALGORITHM

ON CLOUD COMPUTING

Typical scheduling algorithms have different properties, and a specific scheduling decision can support a class of multiple processes. Considering the properties of the different algorithms to picking in a specific circumstance. Several criteria have been proposed, in contrast to booking calculations that decide with the ideal.

5 DISCUSSION

The task scheduling is perhaps these large test into cloud computing. Them purpose of the task scheduling guideline to allow clients to approach virtual machines (VMs) operating times, burden balancing, the standard of services (QoS), reply time with decency assets allotment by remembering various parameters inside which tasks could be presently executed. Inserted some algorithm thinks about simply burden balancing season some think about reply time. While almost all of the algorithm work and a couple of frameworks, expected to which great results cannot be met. Better results may be given by pairing all more and more scheduling measurements to produce an advanced algorithm, although this is a complex one.

6 CONCLUSION

2390

attractive administration to customers and increase the visibility offered through cloud status. The basic purpose of the task scheduling for cloud environments to make less the implementation time of assignments to increase that use of resources. Into this

paper, an examination identified with different occur task scheduling algorithms into the cloud framework is performed. A small depiction of each algorithm philosophy is displayed and most algorithms consider some parameters. The progressively good consequence could be achieved by means of adding further

measurements into occur algorithm. The table depends on the various scheduling framework; For example, loads balancing implementation time the standard of services, reply time with makepan. Serious issues in scheduling algorithms are load adjustment, resource utilization, response time, and memory storage units. An effective scheduling algorithm could be accomplished by means of consolidating various frameworks into an existing algorithm that does make better their general presentation (performance) of the cloud environments.

Table-1 Comparison of scheduling algorithms on cloud computing.

Scheduling Algorithm Methodology Scheduling

Parameters Finding Limitation

First Come First Serve (FCFS) [11]

The algorithm basically controls task scheduling along the FIFO queue. The task that comes initially will be generated first on the VM.

Time of arrival Basic and quick execution

Task scheduling is basically is all about the arrival time, do not consider other criteria. Less usage of VM

Min-Min, Max-Min Algorithm [12],[13],[14]

The algorithm works on a strategy that has a minimum execution time for all tasks

Makespan Enhancedmakespan

Load imbalance and Poor Quality of services

Round Robin [19],[20]

The algorithm depends on a cycling approach where every task has a similar chance of being selected and an equally small unit of time for execution

1. Arrival time 2. Time slice

The time of response is good, balancing the load and fewer complexes.

Causes the pre-elution process once the slice is finished

Priority-Based Job Scheduling Algorithm [26],[27]

Dependency mode Priority to each queue

1. The set of priorities of the process increases with the increase in time.

2. Easy to use and access or user-friendly. access

3. Best for the application that needs time or resources. 4. Less time over

1. The jobs with the lowest priority will be lost when the system crashes.

2. Starvation for the resources they need.

Genetic Algorithm [30],[31]

The algorithm seeks an illustration of the solution domain and an appropriate function to approximate the solution domain.

1. Population size 2. Crossover probability

3. Mutation probability

1. It may be solved through mathematical problems and financial problems more accurately.

2. Simple and very easy to analyze concepts. 3. Few more

applications need very less time of intervals for a processing task.

7 REFERENCES

[1] Zhang, J., Huang, H., & Wang, X. (2016). Resource provision algorithms in cloud computing: A survey. Journal of Network and Computer Applications, 64, 23-42.

[2] Kaur, G., & Sharma, E. S. (2014). Optimized utilization of resources using improved particle swarm Optimization-based task scheduling algorithms in cloud computing. International Journal of Emerging Technology in Advanced Engineering, 4(6), 110-115.

[3] Pasupuleti, S. K., Ramalingam, S., & Buyya, R. (2016). An efficient and secure privacy-preserving approach for outsourced data of resource-constrained mobile devices in cloud computing. Journal of Network and Computer Applications, 64, 12-22. [4] Foster I, Zhao Y, Raicu I, Lu S. Cloud computing and grid

computing 360-degree compared. Grid Computing Environments Workshop, 2008 GCE'08; 2008: IEEE.

[5] Mell P, Grance T. The NIST definition of cloud computing. National Institute of Standards and Technology. 2009; 53(6):50. [6] Rimal BP, Choi E, Lumb I. A taxonomy, survey, and issues of

cloud computing ecosystems. Cloud Computing: Springer; 2010. p. 21-46.

[7] Susanne Albers, ―Energy Efficient Algorithms‖, Communication of ACM, Vol.53, No.5, 2010.

[8] Rajan, A. P. (2013). Evolution of cloud storage as a cloud computing infrastructure service. ArXiv Preprint arXiv:1308.1303.

[9] Guo L, Zhao S, Shen S, Jiang C. Task scheduling optimization in cloud computing based on a heuristic algorithm. Journal of Networks. 2012; 7(3):547±53.

[10]Abdulhamid SM, Latiff MSA, Bashir MB. Scheduling Techniques in on-demand Grid as a Service Cloud: A Review. Journal of Theoretical & Applied Information Technology. 2014; 63(1). [11]Hamscher, V.; Schwiegelshohn, U.; Streit, A.; Yahyapour, R.:

Evaluation of job-scheduling strategies for grid computing. In: International Workshop on Grid Computing, pp. 191–202. Springer, Berlin (2000)

[12]Bhoi, U., & Ramanuj, P. N. (2013). Enhanced max-min task scheduling algorithm in cloud computing. International Journal of Application or Innovation in Engineering and Management, 2(4), 259-64.

[13]Li X, Mao Y, Xiao X, Zhuang Y. An improved max-min task-scheduling algorithm for the elastic cloud. Computer, Consumer and Control (IS3C), 2014 International Symposium on; 2014: IEEE.

[14]Devipriya S, Ramesh C. Improved max-min heuristic model for task scheduling in cloud. Green Computing, Communication and Conservation of Energy (ICGCE), 2013 International Conference on; 2013: IEEE.

[15]Wang G, Yu HC. Task Scheduling Algorithm Based on Improved Min-Min Algorithm in Cloud Computing Environment. Applied Mechanics and Materials. 2013; 303:2429 - 32.

[16]Zhang Y, Xu B. Task Scheduling Algorithm based-on QoS Constraints in Cloud Computing. International Journal of Grid and Distributed Computing. 2015; 8(6):269-80.

[17]Tsai M-Y, Chiang P-F, Chang Y-J, Wang W-J. Heuristic scheduling strategies for linear-dependent and independent jobs on heterogeneous grids. Grid and Distributed Computing: Springer; 2011. p. 496-505.

[18]Chen H, Wang F, Helian N, Akanmu G. User-priority guided Min-Min scheduling algorithm for load balancing in cloud computing. Parallel Computing Technologies (PARCOMPTECH), 2013 National Conference on; 2013: IEEE.

[19]Helmy T. and Dekdouk A. Burst Round Robin: As a Proportional-Share Scheduling Algorithm. IEEE Proceedings of the fourth IEEE-GCC Conference on towards Techno-Industrial Innovations, November 2007, Vol. 14, No.11, pp. 424-428. [20]Mohanty, R., Behera, H., S., Patwari, K., Das, M., R., Dash, M.,

and Sudhashree Design and Performance Evaluation of a New Proposed Shortest Remaining Burst Round Robin (SRBRR) Scheduling Algorithm. American Journal of Applied Science, 2009, Vol. 10, No. 6, pp. 1831-1837.

[21]Yaashuwanth, C. and Ramesh R. Intelligent time slice for round-robin in a real-time operating system. IJRRAS, 2010, Vol. 2, No.2.

[22]Mostafa, S., M., Rida, S., Z. and Hamad, S., H. Finding time quantum of round-robin CPU scheduling algorithm in general computing systems using integer programming, IJRRAS, 2010. [23]Yadav, R., K., Mishra, A., K., Prakash, N. and Sharma, H. An

improved Round Robin scheduling algorithm for CPU scheduling. International Journal on Computer Science and Engineering, 2010, No.02, pp. 1064-1066.

[24]Panda, S., K. and Bhoi, S., K. An Effective Round Robin Algorithm using Min-Max Dispersion Measure. International Journal on Computer Science and Engineering, 2012, Vol. 4, pp 45-53.

[25]Supreeth, S., & Biradar, S. (2013). Scheduling virtual machines for load balancing in the cloud computing platform. International Journal of Science and Research (IJSR), India Online ISSN, 2319- 7064.

[26]Li Yang et al, A New Class of Priority-based Weighted Fair Scheduling Algorithm, Physics Procedia, 33 (2012) 942 – 948 [27]Chtourou, H., & Haouari, M. (2008). A

two-stage-priority-rule-based algorithm for robust resource-constrained project scheduling. Computers & Industrial Engineering, 55, 183-194. [28]Agarwal, D., & Jain, S. (2014). Efficient optimal algorithm of task

scheduling in a cloud computing environment. arXiv preprint arXiv:1404.2076.

[29]Pal, A. J. (2011). Design, development and performance analysis of deadline-based priority heuristic for job scheduling on a grid.

[30]J. W. Ge and Y. S. Yuan, "Research of cloud computing task scheduling algorithm based on improved genetic algorithm," in Applied Mechanics and Materials, 2013, pp. 2426-2429. [31]S. Ravichandran and D. E. Naganathan, "Dynamic Scheduling

of Data Using Genetic Algorithm in Cloud Computing," International Journal of Computing Algorithm, vol. 2, pp. 127-133, 2013.

[32]R. Kaur and S. Kinger, "Enhanced Genetic Algorithm based Task Scheduling in Cloud Computing," International Journal of Computer Applications, vol. 101, 2014.

[33]Z. Zheng, R. Wang, H. Zhong, and X. Zhang, "An approach for cloud resource scheduling based on Parallel Genetic Algorithm," in Computer Research and Development (ICCRD), 2011 3rd International Conference on, 2011, pp. 444-447

[34]S. Singh and M. Kalra, "Scheduling of Independent Tasks in Cloud Computing Using Modified Genetic Algorithm," in Computational Intelligence and Communication Networks (CICN), 2014 International Conference on, 2014, pp. 565-569. [35]V. V. Kumar and S. Palaniswami, "A Dynamic Resource