Introduction

Grid computing is a system where many computers, located in different places, work together to solve large problems. These computers share resources such as processing power, storage, and data through a network. The aim of grid computing is to improve performance by combining many distributed resources into one powerful system.

Key Grid Terms and Concepts

 Resources

Resources are all the components that are shared and used in a grid system. These include computation power (CPU), storage, network, software, and special devices. Resources are important because they provide the ability to perform tasks in the grid.

 Computation

Computation is the process of performing calculations or executing programs. In grid computing, computation is distributed across many computers, so tasks are divided into smaller parts and processed simultaneously to increase speed.

 Storage

Storage refers to the space used to save data, files, and results. In a grid, storage can be distributed across different machines, allowing large amounts of data to be stored and accessed from different locations.

 Communications

Communications refer to the transfer of data between different computers in the grid. This is done through networks such as the internet. Communication is essential because it allows systems to share data and coordinate tasks.

Software and Licenses

Software refers to the applications and programs used to perform tasks in the grid. Licenses are permissions required to use certain software. Some software cannot be used freely, so licenses control access and usage.

Special Equipment, Capacities, Architectures, and Policies

This term includes several important aspects of a grid system:

• Special equipment: advanced machines such as supercomputers or scientific instruments.

• Capacity: the amount of resources available, such as processing power or storage size.

• Architecture: the design and structure of the grid system.

• Policies: rules that control how resources are used, including who can access them and when.

Jobs and Applications

A job is a task submitted by a user to the grid system, such as running a program or analyzing data. An application is the software that performs the job. The grid executes jobs using applications on available resources.

Scheduling

Scheduling is the process of deciding when and where a job will be executed. The scheduler selects suitable resources based on availability, priority, and system policies to ensure efficient execution.

Reservation

Reservation is the process of booking resources in advance. It allows users to guarantee access to specific resources at a certain time, which is useful for important or time-sensitive tasks.

 Scavenging

Scavenging is the process of using idle or unused resources in the grid. When a computer is not being used, the grid can assign it tasks, which improves resource utilization and reduces waste.

Grid Software Components

Grid software components, also known as middleware, are programs that manage the grid system. They handle job submission, resource allocation, communication, and security. Middleware connects users to the resources and ensures everything works together smoothly.

 Intragrid

Intragrid refers to communication and resource sharing within a single grid system, usually within one organization. All components operate under the same management and policies.

Intergrid

Intergrid refers to communication between different grid systems. It allows multiple grids from different organizations to connect and share resources, forming a larger system.

How Major Grid Components Work Together

All grid components work together in a coordinated process to complete tasks efficiently.

First, a user submits a job to the grid system. The grid software (middleware) receives the job and checks the available resources and user permissions. Then, the scheduler decides where and when the job should be executed based on resource availability and system policies.

After scheduling, the grid allocates the necessary resources such as computation power, storage, and network. The job is then sent to the selected machines, where computation takes place. During execution, communication systems transfer the required data between different nodes.

Once the job is completed, the results are stored in storage systems and then sent back to the user. This process ensures that all resources in the grid are used efficiently and tasks are completed faster than using a single computer.