Distributed system is a software system in which components are located on networked computers. They communicate with each other by passing message. Main features of such kind of structure are concurrency of components, lack of a global clock and independent failure of components. It is used to solve computational problems and problem is divided into many subtasks. Each subtask is resolved by one or more computers through message passing. Computational entities or nodes have own local memory and they are independent of others and do share not physically memory or processors. In a distributed system there are two predominant ways of organizing computers which are described as follow:
Client server system has only one single server and lots of clients that communicate with help of it. This structure is appropriate for service-oriented situations in which server respond to client queries. The major drawbacks of this kind of system is that resources become scarce if there are too many clients and with changing demand it cannot grow and eventually shrinks (Parashar, 2000).
Peer to Peer systems: In this approach labor is divided among all the components of the system. All components of model are able to send and receive data and they all contribute some processing power and memory. This structure provides flexibility to system to increase the size and capacity of computational resources.
Parallel systems: In this model, processors are internally linked with each other and multiple processing units are connected via some interconnection network. Shared memory or message passing methods are used to exchange data. Multiple processors are used to increase the speed (Wilikinson, 2011).
Objective: The purpose of the study was to understand the concept of grid computing and its special issues. Project explored the challenges which come in developing, deploying, promoting and use of grid computing. Research report shows the example of an organization in which it describes the precondition and past condition of firm. It identifies the changes that occur when implementing grid computing and mark the benefits that arise in form of improved information storage infrastructure of company.
Grid computing is a special type of parallel computing that combines the computers with multiple administrative domains. In this kind of computing, grids are loosely coupled, heterogeneous, and geographically dispersed. A super virtual computer is used in this model and it is helpful in solving the scientific, mathematical, and academic problems in an effective manner (Boutboul, and et.al. , 2005). This new technology has enabled the clustering of wide distributed resources like data sources, supercomputers, storage system that can be used as a unified resource.
Grid technique is giving great contribution in scientific research in analyzing and storing massive amounts of data. Main vision of grid is to provide computational powers (resources and services data) to the world to make expertise in many diverse areas. Goal of computing is to develop service oriented architecture and allow organization to sharing distributed software, hardware and information resources in significant manner so organization can accomplish its objective (Dabas and Arya, 2013).
Research paper is focused on challenges such as application, formulation & development, resource management, conceptual & implementation, and networking & security etc. that are faced by grid infrastructure at all levels. Other major challenges of grid computing are Scheduling of tasks, Distributed management, efficient algorithms and problem solving methods and Performance analysis and resource monitoring. The report evaluates organization's current information model and understands the solutions provided by grid computing to an enterprise information department of a company (Dabas and Arya, 2013).
Main goal of grid is to control and share uniform computing resources with seamless global aggregation of resources and services. It is focused on clustering of individual systems to increase computational power and increase storage capacity by combining specialized resources in well planned manner. This topic is chosen to be studied because it is helpful for any organization to accelerate business processes and increase productivity & collaboration of resource (Berman, 2003).
Research report is categorized in three parts. In first part we describe the concept of grid computing. In second part we explained the challenges that are faced by grid architecture. Third part of the project explored the Industrial Sector Data Grid. Brief overview of three sections is describes as follow:
Concept of Grid: We understand the perception of grid computing in business and understand the meaning of grid and its benefits for firms. It offers affordable and flexible IT solutions to organization so they can meet their requirements in constantly changing environment. This technology creates a virtual application operating environment where firm can easily share resources and large amount of data store in flexible manner. Grid distributing virtualizes application services execution and contributes in self-managed operating environment. The goal is to provide a better understanding of Multi-core Parallelism and it is addressing a variety of business requirements. The important features of grids are virtual organization, in which user signing on at one location would view other computer at remote location. It can allocate resources to individual tasks as per needs. After completing the tasks resources are returned to the pool. Support of grids provides solutions on respective issues like resource management, resource monitoring, communication and performance (Daniel, 2004).
Challenges faced by Grid architecture: In this part we understand the issues and problems like application formulation & development, resource management, conceptual & implementation, and networking & security etc. This section provides the solution to overcome those troubles ( Joseph and Fellenstein, 2004).
Industrial Sector Data Grid: In this section we evaluate its current information model and understand the solutions provided by grid computing to an enterprise information department of a company. In this part we have taken a case of an organization whose infrastructure is heterogeneous and complex in nature. No strong integration between departments and individual section of the firm and they are working in separated environments. Presently they collect files and store them in network-attached storage (NAS) and storage area network (SAN). Users working in different areas are faced with problems in sharing data across geographically distributed sites and their unplanned storage capacity cannot store large amount of data in flexible manner (Boutboul and et.al. , 2005). Result occurs in form of deteriorating customer service quality and reduced life of existing assets. In this context, grid provides solutions and improves firm's information infrastructure. Grid technology creates a virtual application operating environment where firm can easily transmit and store large amount of data in flexible manner (Parashar, 2000).
From the above study it is clear that grid is very helpful for firms and it provides flexibility to the organization to effectively utilize their resources in well planned manner. Authorization, Authentication, Resource Access and Resource Discovery are four important procedures that come in a distributed computing system. Benefits of grid computing are optimized utilization of underutilized resources, increase capacity and productivity, parallel processing capacity, efficient resource balancing and support heterogeneous system effectively (Berman, 2003).
Although tremendous benefits are provided by grid computing but some barriers and challenges are also present in developing, deploying, promoting and use of this technology. Main challenges are faced such as load balancing, resource monitoring, distributed management, standard protocols , efficient algorithms and problem solving methods, centralized management and performance analysis and resource monitoring etc.
The current architecture of organization contains heterogeneous disks and platforms distributed sites. There is no proper integration between departments such as manufacturing, engineering and accounting. In present situation location 1 and location 2 have their own department. Location 1 department cannot access location 2’s storage. They use different domain such WAN to retrieve information from WEB. In such condition one side plenty of storage takes place while other side there is shortage of storage. Current environment restricts users to access information and share resources across the domains (Boutboul and et.al. , 2005).
On the other hand, the new architecture helps organization and enables users to split information and effectively utilize resources in well planned manner. It provides a way through which heterogeneous and distributed files can be accessed in simpler and secure way.
Globus Toolkit technology is adopted by firm that is helpful for organization and it allows users to share computing power and databases across corporate and geographical boundaries in secure manner without sacrificing local autonomy. The TeraGrid adopted by firm creates fastest distributed infrastructure and enhance their computing power. It offers high-resolution visualization environments that can manage and store nearly 1 petabyte of data efficient manner (Hesham, and El-barr, 2005).
New distributed environment set up for organization is fulfilling the needs of information departments. This technology allows system administrators to share and reallocate storage space basis on needs. It enables engineers to share information across wide area and avoids unplanned storage upgrades. The new information integration solution is based on IBM WebSphere. This product provides facility to organization to access information in both structured and unstructured way. The framework contains features like data and content federation, replication and event publishing. Two servers are placed at location 1 and one server is put at location 2 to increase the working efficiency of system. Storage medium such as internal disk, relational database management systems and SAN are shared effectively. This structure increases transparency and supports different storage media, and operating systems (Boutboul and et.al. , 2005).
Grid consists of middleware infrastructure which integrates geographically distributed computational resources. When complex, multidisciplinary, and multidimensional kind of operation are executed then there is need of distributed grid infrastructure. In Grid computing, node can be independently operated from different locations. It provides the benefits of Virtual resources and virtual organization, easy access of resources of organization from one location area to another remote location area in effective manner. It gives the direction to reputed firm to solve out the computation of high data intensive problems in financial modeling image rendering, earthquake simulation etc. It manages resource pool after allocating the resources when their task is finish then they are returned back to the pool and available for next allocation.
Grid technology gives several benefits to organization but in some area some more improvements are required. Sometime service availability and load balancing type of issue occurs. Present grid architecture is very complex and grid development firm should focused on creates uniform user friendly environment. This techniques faced trouble to examine the database related problem. For that efficient algorithms and problem solving methods should be developed. Due to complex nature of grid management and administration kinds of issue also occurring. There are other challenges such as hidden costs, scope of grid computing, failure in management and administration of resources, performance analysis and resource monitoring. Grid developing company should work out on those issues and shift user attention from grid computing to service computing.
Architecture of organization is discussed in which the present infrastructure when grid is not applied is studied and the affects of implementation of grid on reputed firm are seen. We revised that corporation has been facing several challenges. It is using heterogeneous disks and platforms distributed sites to share data. They cannot use other department data because location 1 and location 2 has own department and lack of integration is there between different departments. In such condition one side user meets plenty of storage while other side shortage of storage type of problems occurs. Current environment restricted user to access information and share resources across the domains.
After that we examined the changes that are taking place due to implementation of grid computing in organization. While the new architecture adopted by firm enables users to split information in wide geographic area with effective utilization of resources in well planned manner. Globus Toolkit technology and TeraGrid kinds of grids technology helps company to create fastest distributed infrastructure and enhance their computing power. Other revolutionizing changes arise in form of design high-resolution visualization environments that can manage and store nearly 1 petabyte of data efficient manner. Two servers are placed at location 1 and one server is put at location 2 and it set up increase the working efficiency of system.
The research report concluded that grid computing plays an important role in success of an organization. From the above study it is clear that grid is very helpful for firms and it provides flexibility to the organization to effectively utilization of their resources in well planned manner. We need security with high availability of data and on-demand provisioning of resources so organization can achieve its goal in significant way. We understand the perception of grid computing in business and understand the meaning of grid and its benefits for firms. This technology accelerated business processes and increase productivity & collaboration of resource in wide geographical area. Grid technology gives several benefits to organization but in some area some more improvements are required. Sometime service availability and load balancing type of issue occurs.