Grid Computing: Technologies And Challenges

Technologies Associated with Grid Computing

Question:

Write research topic about “Grid computing”.

The report is prepared for the providing a detailed description on the grid computing and the technologies associated with the distributed computing. It is used for the management of huge volume of data and securing the organizational resources for the development of a high performance computer system without increasing the cost of the system. There has been an improvement in the performance of the computing system with the development of the high end processor. Research has been done on the distributed computing system for the inclusion of the clusters in the grid computing system (Chang 2015). There are different challenges faced for the management of the servers installed in the distributed computing system and enabling sharing of the data and the resources. Grid computing is thus applied for sharing data parallel and use the resources depending upon the availability or capability of the devices. The developer face problem to design the grid and different products are available for the management of the grid computing and they are discussed in the report.  The grid computing is the future of the distributed computing for the improvement in the data availability and improvement of the productivity of the data. Different protocols are used for the allocation of the resources and the directory service for improvement of the security of the grid computing.

There are different problems faced for the development of the grid computing and they are studied for the estimation of the limits and behavior of the distributed computing models. Different articles and journals are used for the identification of the order of development of the grid computing. Nesmachnow et al. (2013) stated that grid computing can be used for aggregation of the geographically distributed resources and increase the computation power. The grid application mainly depends on the network and the users, data and the service needs to be able to communicate with each other for the improvement of the performance of the grid computing. The network performance should be high for scheduling the jobs and monitoring the jobs performed in the grid computing. The fault tolerance should be enabled such no data are lost in the network and address the problem of data loss in the network. Autonomic systems can be used for the making autonomous decision and adapt the changes in the network and develop a autonomic network aware scheduling infrastructure.

Bazinet, Zwickl and Cummings (2014) used a data grid simulator for the demonstration of the artifacts and essential blocks are used for the simulation of the data and creation of a programming environment for the management of the grid unaware applications. A simple programming interface can be used for the configuration of a peer to peer network and implementation of network overlay for the improvement of the execution efficiency of the grid computing.

Zheng and Veeravalli (2014) proposed a service oriented methodology for designing and implementing visualization system in the grid computing. The traditional dataflow approach is used for the sharing of the memory and virtualization of the web service and linking it to the pipeline utilizing the subscription and enabling notification for Globus Toolkit. A collaborative visualization is used allowing distributed research team for working collaboratively on the visual analysis of the data. A sKML language is used by the author for the development of the gVIZ project for presentation of the service oriented approach and deployment of the visualization approach for the grid computing.

Challenges in managing Grid Computing

Hameed et al. (2016) stated workflow management system including the services and user portal for the management of the workflow and scheduling of the sub workflow. Monitoring, simulation and the execution of the functionalities used at the grid level. A resource management system should be created and a local sub grid should be managed for scheduling the workflow and management of the conflict arising in the system. The author uses a fuzzy timing technique for addressing the challenges for management of the workflow in a dynamic grid and cross domain grid environment. The workflow needs to be optimized for solving the conflicts in the system and increasing the efficiency of the grid computing.

A considerable measure of heterogeneous equipment is utilized as a part of request to make the Grid and, what’s more, these gadgets are not overseen by just a single individual but rather by various framework executives in each of the organizations. Matrix takes after the difficulties that should be made plans to outfit the full energy of lattice:-

No reasonable standard: – Grid registering utilizes different models, yet all matrices are not utilize same norms. Case all matrix working framework, for example, Apache, Linux and My SQL are utilizing UDDI, WSRF, WWW, XML and SOAP guidelines (Jiang and Chen 2015).

Oracle 10g undertaking actualize without WSRF. IBM builds up the Grid middleware in view of J2EE. We cannot utilize distinctive OS at a similar machine in a similar time in network registering.

Conveyed figuring Vs Grid processing: – Grid registering includes dynamic virtual association, asset sharing and distributed figuring (Shojafar et al. 2015). The Grid plans to make access to figuring power, logical information archives and trial offices as simple as the Web makes access to data. Same all offices give the matrix registering. So it is a test for framework figuring.

Absence of network empowered programming: – The product, which are empowered the framework figuring are less, It has restricted programming on Grid. Much programming has not copyright issues and source code of permit. It is requirement for more organization creating matrix empowered adaptation, require more engineers on network improvement and need to create open source programming (Bazinet, Zwickl and Cummings 2014).

Sharing Resources between Various Types of Services: – Grid utilized for sharing asset from different locales and framework has. It handles an enormous measure of information as a framework stage. A ton of destinations and numerous servers assembled there it is so mind boggling framework (Shamshirband et al. 2013). It gives trouble to equipment asset sharing inside virtual association.

Hard to create: – Grid programming utilized java, XML, utilize web administrations UDDI, WSRF, WSDD, WSDL and GT3 developing rules. It is an issue who building up the matrix Applications. Essentially that is accessible for senior software engineering designers and undertaking designers.

In spite of the fact that the execution of every usage can be believed to rely upon the information sharing and correspondence example of the application program, some broad patterns can be watched (Arabnejad and Barbosa 2014). It is discovered that utilizing various procedures on SMP hubs gives great speedups just in programs that have almost no information sharing and correspondence. In every single other case, the quantity of page flaws is high, and causes overabundance correspondence. Numerous application strings can enhance the execution now and again, by decreasing the quantity of page shortcomings (Hameed et al. 2016). This is extremely powerful when there is an expansive level of sharing over the strings among the nodes connected in the distributed environment. The utilization of client level threads results in expansion of computation and response time, since every one of the strings seek CPU time on a solitary processor (Gong et al. 2015). On the off chance that part strings are utilized furthermore, the general execution enhances essentially in every one of the projects tried. Utilizing a devoted correspondence string to survey for approaching messages is a favored contrasting option to flag driven I/O. The simultaneous dsm server approach lessens the page fault latency while enabling various solicitations to be taken care of simultaneously (Setia and Jain 2016). Lastly, utilizing a various leveled summing boundary enhances the obstruction hold up times in a large portion of the projects.

Data Grid Simulator

Restricted Area and Applications:- Grid registering is utilized to take care of for vast and complex issues. Its region is restricted, for example, researchers, engineer, examination and scientists. In general, utilizing part strings is exceptionally encouraging, particularly for normal projects with minimal false sharing. Extra work should be done to distinguish the constraint of the grid computing, since this rules the execution time in the situations where the general outcomes are not that great (Gokuldev, Ashokan and Rajeev 2016). The main of the network development team is on the improvement of the network performance and most of the time the consistency of the network is overlooked.

Biometric research, Hollywood flicker and the industrial research finds a great space for the application of the grid computing for sharing of the resources and it is utilized for the management if the constraints.

Administration and management: – Many foundations and associations utilized framework processing. It disseminates the assets on extensive geologically circulated situations and gets to the heterogeneous gadgets and machines (Smanchat and Viriyapant 2015). So it is a noteworthy test to deal with the organization of the matrix registering.

For some, expansive undertakings, registering of the framework is the essential answer for quickening the development of the grid computing environment and include different distributed computing environment for sharing of the data between the systems. Also, for network empowered applications, augmenting execution and scale are the essential concerns. In any case, not all framework middleware is the same (Arabnejad, Barbosa and Prodan 2016). A few items force structural constraints or confine your decision of working frameworks or engineer instruments. Another potential imperative is influencing various lines of business (LOBs) to share a typical foundation (Lin, Chen and Chang 2013). Their dread of losing control and missing administration level goals can prompt unwieldy, costly, application particular frameworks estimated to crest request.

A comparison is made on the ORACLE and IBM grids depending on the different factors and development of a highly integrated environment of grid.

There are different workflow scheduling algorithm that can be used for the management of the interdependent task for grid computing. The task are designed for the cluster environment and the resources are shared in the cluster environment for competing the resources. The workflow should be schedules because the resources cannot be controlled by the scheduler and all the resources may not perform identically for a task (Citeseerx.ist.psu.edu. 2017). The applications used in the grid computing are data intensive and a huge volume of data sets needs to be transferred between the different nodes in the network and thus the following workflow scheduling algorithm are applied for addressing the issues for management of the huge volume of data between different sites.

The grid computing protocol can be classified into five different categories such as:

• Grid network communication and grid data transfer protocol
• Grid information security protocol
• Grid resource information protocol
• Grid management protocol, and
• Grid interface protocol

The following figure is used for classification of the protocol and given in the below figure:

The protocols are used for different layers such as resource discovery protocol is used for application collective layer along with co allocation protocol, SLA and grid monitoring protocol. The resource registration protocol and secure communication protocol is used for securing the network and TCP or UDP protocol is used for enabling communication between the nodes connected in the network.

Service-Oriented Methodology

The future architecture of the grid computing is based upon the standard base and interoperable interfaces. The future architecture is designed for providing a standard interface for each of the layers of the hourglass model and utilizes the OGSA interface for the different core grid service standardized by global grid forum (Valentini et al. 2013). The resource should be allocated in advance for using it in future data. The future of the grid computing depends upon the service oriented architecture, generalized resource management for the management of the workloads between the nodes connected in the network (Plank and Thomason 2012). Best effort service can be provisioned and service level agreement can be managed for the addressing of the requirement of the organizations.

Conclusions

Grid computing has become a platform for the management of the distributed data and different grid software and stacks are used for the management such as Globus Toolkit. The grid technology has improved and it works on the same principle of power grids and thus the computing service gest improved. The grid computing can be used for decreasing the cost of the usage of the computer resources and used for solving different problems that cannot be solved with the application for high computation power. The grid computing reduces the chances of wastage of the resources by using an ideal system for other computation task. Different service oriented designs are proposed for the development of a grid environment and different technology such as the virtualization, gateway server and scheduling the applications can be used for the development of the grid environment. The different features and architecture of grid computing is used for the identification of the opportunities and challenges faced in grid computing. The participants are included in the analysis for controlling the decision process and finding the algorithm suitable for the finding the change in the state of transition of the network. The centralized and the distributed system are compared with each other and programming models are analyzed for expansion of the network. A large volume is data is managed in grid computing and different toolkits are used for the management of the resources in the organization.

References

Arabnejad, H. and Barbosa, J.G., 2014. A budget constrained scheduling algorithm for workflow applications. Journal of Grid Computing, 12(4), pp.665-679.

Arabnejad, H., Barbosa, J.G. and Prodan, R., 2016. Low-time complexity budget–deadline constrained workflow scheduling on heterogeneous resources. Future Generation Computer Systems, 55, pp.29-40.

Bazinet, A.L., Zwickl, D.J. and Cummings, M.P., 2014. A gateway for phylogenetic analysis powered by grid computing featuring GARLI 2.0. Systematic biology, 63(5), pp.812-818.

Bazinet, A.L., Zwickl, D.J. and Cummings, M.P., 2014. A gateway for phylogenetic analysis powered by grid computing featuring GARLI 2.0. Systematic biology, 63(5), pp.812-818.

Chang, J.Y., 2015. A Distributed Cluster Computing Energy-Efficient Routing Scheme for Internet of Things Systems. Wireless Personal Communications, 82(2), pp.757-776.

Citeseerx.ist.psu.edu. (2017). [online] Available at: https://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.269.2371&rep=rep1&type=pdf [Accessed 24 Dec. 2017].

Gokuldev, S., Ashokan, A. and Rajeev, R., 2016. A DTQ Scheduling Algorithm with Check pointing approach in Computational Grid. International Journal of Applied Engineering Research, 11(9), pp.6850-6855.

Gong, Y.J., Chen, W.N., Zhan, Z.H., Zhang, J., Li, Y., Zhang, Q. and Li, J.J., 2015. Distributed evolutionary algorithms and their models: A survey of the state-of-the-art. Applied Soft Computing, 34, pp.286-300.

Hameed, A., Khoshkbarforoushha, A., Ranjan, R., Jayaraman, P.P., Kolodziej, J., Balaji, P., Zeadally, S., Malluhi, Q.M., Tziritas, N., Vishnu, A. and Khan, S.U., 2016. A survey and taxonomy on energy efficient resource allocation techniques for cloud computing systems. Computing, 98(7), pp.751-774.

Hameed, A., Khoshkbarforoushha, A., Ranjan, R., Jayaraman, P.P., Kolodziej, J., Balaji, P., Zeadally, S., Malluhi, Q.M., Tziritas, N., Vishnu, A. and Khan, S.U., 2016. A survey and taxonomy on energy efficient resource allocation techniques for cloud computing systems. Computing, 98(7), pp.751-774.

Jiang, Y.S. and Chen, W.M., 2015. Task scheduling for grid computing systems using a genetic algorithm. The Journal of Supercomputing, 71(4), pp.1357-1377.

Lin, J.W., Chen, C.H. and Chang, J.M., 2013. QoS-aware data replication for data-intensive applications in cloud computing systems. IEEE Transactions on Cloud Computing, 1(1), pp.101-115.

Nesmachnow, S., Dorronsoro, B., Pecero, J.E. and Bouvry, P., 2013. Energy-aware scheduling on multicore heterogeneous grid computing systems. Journal of Grid Computing, 11(4), pp.653-680.

Plank, J.S. and Thomason, M.G., 2012. Processor allocation and checkpoint interval selection in cluster computing systems. Journal of Parallel and distributed Computing, 61(11), pp.1570-1590.

Setia, H. and Jain, A., 2016, July. Literature survey on various scheduling approaches in grid computing environment. In Power Electronics, Intelligent Control and Energy Systems (ICPEICES), IEEE International Conference on (pp. 1-4). IEEE.

Shamshirband, S., Anuar, N.B., Kiah, M.L.M. and Patel, A., 2013. An appraisal and design of a multi-agent system based cooperative wireless intrusion detection computational intelligence technique. Engineering Applications of Artificial Intelligence, 26(9), pp.2105-2127.

Shojafar, M., Javanmardi, S., Abolfazli, S. and Cordeschi, N., 2015. FUGE: A joint meta-heuristic approach to cloud job scheduling algorithm using fuzzy theory and a genetic method. Cluster Computing, 18(2), pp.829-844.

Smanchat, S. and Viriyapant, K., 2015. Taxonomies of workflow scheduling problem and techniques in the cloud. Future Generation Computer Systems, 52, pp.1-12.

Valentini, G.L., Lassonde, W., Khan, S.U., Min-Allah, N., Madani, S.A., Li, J., Zhang, L., Wang, L., Ghani, N., Kolodziej, J. and Li, H., 2013. An overview of energy efficiency techniques in cluster computing systems. Cluster Computing, pp.1-13.

Zheng, Q. and Veeravalli, B., 2014. On the design of mutually aware optimalpricing and load balancing strategiesfor grid computing systems. IEEE Transactions on Computers, 63(7), pp.1802-1811

Order an Essay Now & Get These Features For Free:

Turnitin Report

Formatting

Title Page

Citation

Outline

Place an Order