Study on Computer Generated Electromagnetic Effects on Computer Users

121

Copyright © 2011-15. Vandana Publications. All Rights Reserved.

Volume-5, Issue-2, April-2015 International Journal of Engineering and Management Research

Page Number: 121-125

Fictional Simulation

Shehin Shams P1, Suchithra M B2, Swathy V S3, Vineesha K V4, Mrs Nitha K P5 1,2,3,4 Department of CSE, Vidya Academy of Science and Technology, Thrissur, INDIA Assistant Professor, Department of CSE, Vidya Academy of Science and Technology, Thrissur, INDIA

ABSTRACT

Grid technologies have progressed towards a service-oriented paradigm that enables a new way of service provisioning based on utility computing models. One of the most challenging problems in grid environment is workflow scheduling. Appropriate scheduling algorithm is selected by workflow management systems. Different scenarios require different scheduling algorithms. The selection of a particular scheduling algorithm depends upon various factors like the parameter to be optimized (cost or time), quality of service to be provided and information available regarding various aspects of job. In this survey, we investigate existing workflow scheduling algorithms.

I.

INTRODUCTION

Grid computing, also called computational grid, was developed by computer scientists in the mid-1990s based on the inspiration of the electrical power Grid’s pervasiveness, ease of use, and reliability, to provide a computational power grid infrastructure for wide-area parallel and distributed computing. The motivation for computational Grids was initially driven by large-scale, resource (computational and data) intensive scientific applications that require more resource than a single computer (PC, workstation, supercomputer, or cluster) could provide in a single administrative domain.

As a computing infrastructure, a Grid enables the sharing, selection, and aggregation of a wide variety of geographically distributed resources owned by different organizations for solving large-scale resource intensive problems in various fields. In order to build a Grid, the development and deployment of a number of services is required. They include low-level services such as security, information, directory, resource management (resource trading, resource allocation, quality of services) and high-level services for application development, resource management and scheduling (resource discovery, access cost negotiation,

resource selection, scheduling strategies, quality of services, and execution management). Among them, the two most challenging aspects of Grid computing are resource management and scheduling. A group of Quality-of-Service (QoS) driven algorithms are presented for the management of resources and scheduling of applications.

There are several types of grids available such as community grids, data grids, utility grids and so on. Table 1 shows some differences between community Grids and utility Grids in terms of availability, Quality of Services (QoS) and pricing. In utility Grids, users can make a reservation with a service provider in advance to ensure the service availability, and users can also negotiate with service providers on service level agreements for required QoS. Compared with utility Grids, service availability and QoS in community Grids may not be guaranteed. However, community Grids provide free access, whereas users need to pay for service access in utility Grids. In general, the service pricing is based on the QoS level and current market supply and demand.

122

Copyright © 2011-15. Vandana Publications. All Rights Reserved.

focuses on a combination of these parameters.Workflow is usually used to represent by the directed acyclic graphs (DAG).In majority of the grid environments recently used the Simgrid tool as simulator.It is based on java programming language.grid sim was used before that.it was developed using basic c programming language.Due to it’s some of the limitations Simgrid developed.

TABLE 1. COMMUNITY GRIDS vs. UTILITY GRIDS

II.

WORKFLOW SCHEDULING

In workflow scheduling, different sub tasks of a bigger task are allocated resources in such a way that some pre-defined objective criteria is met. There are various problems in bioinformatics, astronomy and business enterprise in which a set of sub tasks is executed in a particular sequence in order to carry out a bigger task. In general, a workflow application requires series of steps to be executed in a particular fashion. These steps have parent child relationship. The parent task should be executed before its child task. The parent task is linked to child task according to set of rules. A workflow application is generally represented as a Directed Acyclic Graph (DAG) such as G (V, E) where V is the number of tasks and E is the information regarding data dependencies among tasks. A task which does not have any parent task is called entry task and a task which does not have any child task is called an exit task.

Figure 1 shows the dependencies among different tasks in a workflow graph G. The parent task 0 is executed before child tasks 1, 2, 3 and 4.The output of parent node acts as an input to child node. The task 0 acts as entry node and task 9 act as an exit node. Task 9 is execute after the completion of tasks 5, 6, 7and 8.

In workflow scheduling, the different tasks are allocated resources (e.g. virtual machines). The workflow scheduling decisions are taken by workflow management systems (WfMS), which works as a broker between users and grid service providers (GSPs). Whenever the WfMS accepts a workflow, it contacts Grid information like the Grid Market Directory (GMD), to query about available services for each task and their QoS attributes. Each GSP has to register itself and its services with the GMD, so that it can present and sell its services to users. Then, the WfMS directly contacts the desired GSPs to query about the free time slots of the suitable services. Using this information, the WfMS can execute a scheduling algorithm to map each task of a workflow to one of the available services. According to the generated schedule, the WfMS contacts GSPs to make advance reservations of

selected services. This results in an SLA between the WfMS and the GSP specifying the earliest start time (EST), the latest finish time (LFT), and the price of the selected service. Usually, the SLA contains a penalty clause in case of violation of the service level to enforce service level guarantees.

Fig.1. A Workflow represented in the form of a graph

III.

SURVEY OF WORKFLOW

SCHEDULING ALGORITHMS FOR GRID

COMPUTING

Many heuristics [1] have been developed to schedule inter-dependent tasks in homogenous and dedicated cluster environments. However, there are new challenges for scheduling workflow applications in a Grid

environment, such as:

• Resources are shared on Grids and many users compete for resources.

• Resources are not under the control of the scheduler.

• Resources are heterogeneous and may not all perform identically for any given task.

• Many workflow applications are data-intensive and large data sets are required to be transferred between multiple sites.

Therefore, Grid workflow scheduling is required to consider non-dedicated and heterogeneous execution environments. It also needs to address the issue of large data transmission across various data communication links. The input of workflow scheduling algorithms is normally an abstract workflow model which defines workflow tasks without specifying the physical location of resources on which the tasks are executed.

There are two types of abstract workflow model, deterministic and non-deterministic. In a deterministic model, the dependencies of tasks and I/O data are known in advance, whereas in a non-deterministic model, they are only known at runtime.

The scheduling algorithms are used by WfMS to find optimal map of workflow tasks and grid resources (virtual machines). The role of workflow scheduling algorithm is to find the schedule which satisfies user’s objectives. Users define their objectives in SLA (Service Level Agreement) document which is written between a grid user and a grid service provider. The user may require multiple objectives to be satisfied such as cost optimization, makespan optimization,

Community Grids

Utility Grids

Availability Best effort Advanced reservation

QoS Best effort Contract/SLA

Pricing Not considered or

free access

Usage, QoS level, Market

123

Copyright © 2011-15. Vandana Publications. All Rights Reserved.

is the role of scheduling algorithm to find the optimal schedule which satisfies user’s objectives.

Generally there are two category of the scheduling algorithm; Static Scheduling and Dynamic Scheduling. In Static Scheduling, Tasks are arrives simultaneously and available resource schedule updated after each task is schedule. In Dynamic Scheduling, task and machine set location and allocation is not going to fix. Dynamic strategy applied in two fashions: On-line mode heuristic scheduling and Batch mode heuristic scheduling. In on-line mode heuristic scheduling, tasks are scheduled when they arrived in the system. In Batch mode, tasks are queued and collected into set when they arrive in the system. The scheduling will start after a fixed period of time. In another way two major types of workflow scheduling are; best-effort based and QoS constraint based scheduling. The best-effort based scheduling attempts to minimize the execution time ignoring other factors such as the monetary cost of accessing resources and various users’ QoS satisfaction levels. On the other hand, QoS constraint based scheduling attempts to minimize performance under most important QoS constraints, for example time minimization

under budget constraints or cost minimization under deadline constraints.

TABLE 1: A BRIEF DESCRIPTION AND COMPARISON AMONG VARIOUS WORKFLOW SCHEDULING

ALGORITHMS

Fig. 2. A taxonomy of Grid workflow scheduling algorithms

Scheduling Algorithm Scheduling Type

Scheduling

Parameters Scheduling Factors Finding

Environ

ment Tools

QoS Guided Min-Min

Heuristic [2] Batch Mode

Quality of service, Make span Quality of service,Make span

Bandwidth of tasks

1. Reduce the Makespan then Min-Min

2. Use only bandwidth parameter for QoS

Grid GridSim

QoS Priority Grouping

Algorithm[3] Batch Mode

Acceptance rate,

completion time Grouped tasks

1. Deadline and acceptance rate of the tasks

2. Makespan

Grid GridSim

Towards Improving QoS-Guided Scheduling[4]

Batch Mode Makespan Grouped tasks[Jobs]

1. Improving makespan to achieve better performance

2. Reduce the Resource Need by Rescheduling

Grid GridSim

QoS based predictive Max-min, Min-min switcher[5]

Batch Mode Makespan Heuristic Better performance with

QoS Grid GridSim

RASA[6] Batch Mode Make span Grouped tasks Use to reduce the

makespan Grid GridSim

HEFT workflow scheduling Algorithm[7]

List scheduling dependency mode

Makespan Highest upward

rank

Reduce make span in a

DAG Grid GridSim

Cost based scheduling on utility grids.[8]

Budget

constrained Cost Task Scheduling

Reschedule the

unexpected tasks Grid GridSim

Task duplicationbased scheduling Algorithm for Network of Heterogeneous systems (TANH)[9]

Duplication based heuristic mode

makespan DAG scheduling Reduced makespan Grid GridSim

Workflow with budget constraints[10]

Budget

constrained Makespan,budget DAG scheduling

Minimize the execution

time and the make span Grid GridSim

Ant colony Optimization Based Workflow

Scheduling[11]

Meta-heuristic Resource

Utilization,time QOS

Optimizes the service

flow scheduling Grid GridSim

Selective Rescheduling

policy[12] Static heuristic Cost and makespan

Minimal spare time and the slack

Selectively reschedule

124

Copyright © 2011-15. Vandana Publications. All Rights Reserved.

Predict Earliest Finish

Time (PEFT)[13] List scheduling time

Scheduling Length Ratio, Efficiency, pair-wise comparison of the number of occurrences of better solutions and Slack.

Reduced scheduling time Grid GridSim

Genetic algorithm[14] Meta-heuristic Deadline,budget QoS Reduced timeand cost Grid GridSim

List Scheduling [15] Heuristics Makespan, Load

Balance DAG scheduling

Optimized makespan and

load balance Grid GridSim

Genetic algorithm[16] Meta-Heuristic Budget Constrained QoS

Minimizes the execution time while meeting a specified user budget

Grid GridSim

DCP (Dynamic Critical

Path) [17] Heuristic Resource Availability Priority of tasks

Better performance where resource availability changes frequently

Grid GridSim

Improved Critical Path using Descendant Prediction (ICPDP) algorithm [18]

Hybrid-heuristic Makespan and load

balance Available resources

Makespan minimization and improve the utilization of resources.

Grid GridSim

Particle Swarm

Optimization[19] Meta-Heuristic

Makespan, Cost and

Reliability Grouped Tasks

Minimizes execution time, cost, and maximize the reliability.

Grid GridSim

Genetic Algorithm[20] Meta-Heuristic Makespan and Cost

Optimization QoS

Multi-Objective Differential Evolution (MODE) that optimize both cost and makespan for workflow application.

Grid GridSim

Particle Swarm Optimization- Rotary Hybrid Discrete Particle Swarm Optimization (RHDPSO) algorithm [21]

Meta-Heuristic Makespan, Cost and

Load Balance Grouped Tasks

Optimize the makespan, cost and perform load balancing when scheduling workflow application.

Grid GridSim

Novel DBC (Deadline and Budget

Constrained) [22]

Heuristic Deadline and Budget

Constrained Task Scheduling

Novel DCP is compared with DCP. The

experiment results show that the workflow completion ratios of Novel DCP are higher than DCP.

Grid GridSim

HGreen Algorithm[23] Heuristic Energy Efficient

Schedules the heavier tasks on maximum green resources..

The simulation results have shown that the H-Green algorithm reduce the power consumption in global grids

Grid GridSim

List Scheduling

Algorithm[24] Heuristic

Makespan, Economic Cost, Energy Consumption, Reliability

DAG scheduling

It outperform as compared with bi-criteria heuristic and bi-criteria genetic algorithms

Grid GridSim

PCP (Partial Critical

Path) [25] Heuristic

Deadline-Constraint,

Cost Minimization QoS

PCP algorithm

minimizes the execution time while meeting the user defined deadline.

Grid GridSim

125

Copyright © 2011-15. Vandana Publications. All Rights Reserved.

: Tick sign means that work has already been done in that

area and there is a workflow scheduling algorithm for solving that type of problem.

? : Question mark sign means that there is a need to explore workflow scheduling algorithm for that particular domain focusing on different aspects like cost optimization, deadline constrained, budget constrained, reliability, load balance, availability and energy efficient.

IV.

CONCLUSIONS

In this paper, we surveyed various existing workflow scheduling algorithms and tabulated them on the basis of nature of scheduling algorithm, type of algorithm, objective criteria and the environment to which the workflow scheduling algorithm was applied. From the literature reviewed, it is clear that lot of work has already been in the area of workflow scheduling but still there are many areas which require further attention e.g. there is a need to explore energy efficient genetic algorithm for workflow application whereas cost and deadline constraints have already been addressed using genetic algorithms.

REFERENCES

[1] Y. K. Kwok and I. Ahmad, “Static Scheduling Algorithms for Allocating Directed Task Graphs to Multiprocessors”, ACM Computing Surveys, 31(4):406-471, Dec.1999

[2] XiaoShan He,Xianhe Sun and Gergor von Laszewski.,“QoS guided Min-Min heuristic for grid task scheduling”, Journal of Computer Science and Technology, 18(4), p.442-451, 2003.

[3] Dong. F, Luo. J, Gao. L and Ge. L, “A Grid Task Scheduling Algorithm Based on QoS Priority Grouping,” In the Proceedings of the Fifth International Conference on Grid and Cooperative Computing (GCC’06), IEEE, 2006.

[4] Ching-Hsien Hsu,Zhan. J.,Wai-Chi Fang, et al. “Towards improving QoS-guided scheduling in grid” ,Third ChinaGrid Annual Conference(CHINAGRID), Dunhuang, Gansu, China, , p.89-9, 2008.

[5] M.Singh and P.K.Suri; “QPSMax-Min<>Min-Min : A QoS Based Predictive Max-Min, Min-Min Switcher Algorithm for Job Scheduling in a Grid”, International Technology Journal7(8) :p.1176-1181,2008 .

[6] Saeed Parsa and Reza Entezari-Maleki,” RASA: A New Task Scheduling Algorithm in Grid Environment” , World Applied Sciences Journal 7(Special Issue of Computer & IT): 152-160, 2009

[7] Wieczorek, M., Prodan, R. and Fahringer, T. ‘‘Scheduling of scientificworkflows in the ASKALON grid environment’’, SIGMOD Rec., 34(3), pp.56–62 (2005).

[8] Yu, J., Buyya, R. and Tham, C.K. ‘‘Cost-based scheduling of scientific wokflow applications on utility grids’’, First Int’l Conference on e-Scienceand Grid Computing, Melbourne, Australia, pp. 140–147 (2005).

[9] R. Bajaj and D.P. Agrawal, “Improving Scheduling of Tasks in aHeterogeneous Environment,” IEEE Trans. Parallel and Distributed Systems, vol. 15, no. 2, pp. 107-118, Feb. 2004 .

[10] Sakellariou, R., Zhao, H., Tsiakkouri, E. and Dikaiakos, M.D. ‘‘Scheduling workflows with budget constraints’’, In Integrated Research in GRID Computing, S. Gorlatch and M. Danelutto, Eds Springer- Verlag., pp.189–202, (2007). [11] W.N. Chen and J. Zhang, “An Ant Colony Optimization Approach to Grid Workflow Scheduling Problem with Various QoS Requirements,” IEEE Trans. Systems, Man, and Cybernetics,vol. 39, no. 1, pp. 29-43, Jan. 2009.

[12] R. Sakellariou and H. Zhao, “A Low-Cost Rescheduling Policy forEfficient Mapping of Workflows on Grid Systems,” ScienceProgramming, vol. 12, pp. 253-262, Dec. 2004. [13] Hamid Arabnejad and Jorge G. Barbosa,” List Scheduling Algorithm for Heterogeneous Systems by an Optimistic Cost Table”,IEEE,2008.

[14] J. Yu and R. Buyya, “Scheduling Scientific Workflow Applications with Deadline and Budget Constraints Using GenetiCc Algorithms,” Scientific Programming, vol. 14, nos. 3/4, pp. 217-230, 2006.

[15] A. Mandal, K. Kennedy, C. Koelbel, G. Marin, J. Crummey and B. Liu,” Scheduling Strategies for Mapping Application Workflows onto the Grid.”, High Performance Distributed Computing, 14th IEEE International Conference,2005.

[16] Jia Yu and Raj Kumar Buyya.” A Budget Constrained Scheduling of Workflow Applications on Utility Grids using Genetic Algorithms.” Workflows in Support of Large-Scale Science, IEEE Conference, Pg. 1-10, 2006.

[17] M. Rahman, S. Venugopal and R. Buyya. “A Dynamic Critical Path Algorithm for Scheduling Scientific Workflow Applications on Global Grids.” E-Science and Grid Computing, IEEE International Conference, 2007Pg. 35-42. [18] Bogdan Simion, Catalin Leordeanu, Florin Pop and Valentin Cristea.”A Hybrid Algorithm for Scheduling Workflow Applications in Grid Environments”. OTM Confederated International ConferencesPg. 1331-1348, 2007. [19] Fli Tao, Dongming Zhao, Yefa Hu and Zude Zhou. “Resource Service Composition and Its Optimal Selection Based on Particle Swarm Optimization in Manufacturing Grid System.” Industrial Informatics, IEEE Transactions, Pg. 315-327,2008.

126

Copyright © 2011-15. Vandana Publications. All Rights Reserved.

Yang Yu. “QoS Constrained Grid Workflow Scheduling Optimization Based on a Novel PSO Algorithm”, Grid and Cooperative Computing. 8th IEEE International Conference, Pg. 153-159, 2009.

[22] Yong Wang, R. M. Bhati and M. A. Bauer. ,”A Novel Deadline and Budget Constrained Scheduling Heuristic for Computation Grids”. Journal of Central South University of Technology Vol. 18, Issue 2, Pg. 465-472, 2011..

[23]R.Bajaj, “Workflow Scheduling Algorithm for Optimizing Energy Efficient Grid Resources Usage, Dependable. Automic and Secure Computing”, 9th IEEE International Conference, Pg. 642-649.

[24] H. M. Fard, R. Prodan, J. J. D Barrionuevo and T. Fahringer.” A Multi-Objective Approach for Workflow Scheduling in Heterogeneous Environment”,. Cluster, Cloud and Grid Computing 12th IEEE International Conference, Pg. 300-309, 2012.