Sample Computer Science Paper on Distributed Computing

Distributed Computing

Introduction

A cluster of computers using distributed computing (also known as distributed processing) connects across a network to exchange data and coordinate processing power. It is called a “distributed system” when it is made up of several computers. Because of its flexibility, scalability, performance, redundancy, and cost-effectiveness, distributed computing is becoming more popular (through the use of low-cost commodity hardware).In database and application architecture, distributed computing has become highly prevalent as data quantities have risen and performance expectations have increased. To make room for growing data quantities, it’s beneficial to expand the system with additional hardware. Traditional “big iron” settings, made up of powerful computer servers, need hardware upgrades and replacements to keep up with increasing demand. In its most basic definition, distributed computing is sharing work across several computers located in various locations worldwide. For the communication and coordination of the activities to go successfully, the distributed system being employed here must be implemented on networked computers.

In general, the primary purpose of distributed computing is to link users with resources to maximize performance while keeping costs as low as possible. The system’s design ensures that even if one of its components fails, the system continues to operate, achieving the expected outcomes. Hence this research gives an in-depth explanation of distributed cloud computing architecture

 

The Following Are Some Examples Of Distributed Systems.

Networks

Networks like LAN and ethernet were initially proposed in the 1970s, and the first distributed systems were built. When computers were able to communicate with each other using local IP addresses for the first time, Email and the Internet as we know it now are two examples of distributed systems that have changed through time. LAN-based distributed systems developed into “Internet-based” distributed systems when the IPv4 address space transitioned to IPv6 address space. ( Snezhko et al..,2010)

Telecommunication networks

Telephone and cellular networks are instances of dispersed networks, as well. – Peer-to-peer networks have been present for more than a century, and the telephone was one of the first examples. In cellular networks, base stations are located in cells, and the networks are dispersed. As VOIP (voice over IP) networks have progressed, they have become more sophisticated as a distributed system.

Distributed Database Systems

In a distributed database, data is spread over numerous servers and geographic locations. Replicating or duplicating data between systems is a possible option.

Distributed databases are common in many popular applications. Thus it’s essential to be aware of the system’s homogeneity or heterogeneity.

Distributed databases that use the same database management system and data model are referred to be “homogeneous.”Adding more nodes and locations makes it simpler to maintain and grow performance.

Heterogeneous distributed databases may accommodate multiple data models and database management systems. Applications and systems merging often create gateways, which are used to transfer data across nodes.

Parallel Processing

The terms “parallel computing” and “distributed systems” used to be separate. In parallel computing, many threads or processors were able to access the same data and memory simultaneously. Each machine has its CPU and memory, making distributed systems more complex. Parallel processing is now a part of distributed computing because of the growth of current operating systems, processors, and cloud services.

Distributed artificial intelligence

The use of large-scale computing capacity and parallel processing enables distributed artificial intelligence (DAI) to handle extensive data sets utilizing multi-agents.

Distributed computing in cloud computing

Cloud computing options and providers have made distributed computing even more accessible. Cloud computing instances do not support distributed computing out of the box. To use readily available computer resources, various distributed computing software applications may be operated in the cloud. DBAs and technology suppliers used to be the only people enterprises could turn to for connecting networks inside and outside of data centers to share resources. Cloud service providers simplify adding servers to a cluster to boost storage capacity or processing speed. ( Hajibaba et al ..,2014)

Because of the simplicity and speed with which additional computing resources may be deployed, distributed computing allows for better degrees of agility when dealing with expanding workloads than traditional computing methods. This allows for “elasticity,” in which a cluster of computers may be readily extended or reduced in response to the current workload demands. In the same way that distributed computing is based upon, cloud computing is also built upon. Technically speaking, if you have an application that syncs information across multiple of your devices, you are engaging in cloud computing. Since it is doing so using distributed computing, you are benefiting from the benefits of distributed computing.

Distributed System Architecture

Distributed systems can’t communicate without a network that connects all of their components (machines, hardware, or software) and allows messages to be sent back and forth. It is possible to connect network components via IP addresses, cables, or even circuit boards. Databases, objects, and files are only some forms of data that computers use to interact with one another. To ensure the integrity of a distributed system’s message delivery and reception is essential to its success, as is how it is accepted. As services and applications expanded in popularity, more computers were deployed and maintained to keep up with demand, so distributed systems were formed. One of the most important trade-offs to consider when designing distributed systems is the trade-off between complexity and performance.

To accomplish a particular purpose, components in a distributed architecture might work together across a communication network that connects them. There are no single machines in this design, but instead, multiple separate computers perform the same functions. There are several ways to show the concept of a distributed system, including client-server architecture, broker architecture, and Service-Oriented Architecture (SOA) (SOA).

.NET, J2EE, CORBA,.NET Web services, AXIS Java Web services, and Globus Grid services are a few of the technological frameworks that allow distributed systems. Distributed application development and execution rely on middleware as a foundation. It acts as a stopgap between the network and the applications. It acts as a middleman between the many parts of a distributed system, overseeing and providing assistance for them. Transaction processing displays, data converters, and communication controllers are a few examples of this technology in use. ( Jo et al..,2014)

Key Advantages Of Distributed Computing

Because they are all connected through a network, the computers in a cluster may act as if they were one. This multi-computer paradigm, despite its complexity, has several advantages:( Jadeja et al ..,2012)

Scalability. The “scale-out design” of distributed computing clusters makes it simple to add extra hardware to accommodate increased workloads (versus replacing existing hardware).

Performance. Using a divide-and-conquer strategy, the cluster may reach high-performance levels via parallelism, whereby each computer simultaneously addresses a part of the entire work.

Resilience. Replicating data over several computer servers ensures that no single point of failure exists in distributed computing clusters. There are backups in case a machine fails, ensuring that no data will be lost.

Cost-effectiveness. Initial cluster deployments and expansions may be relatively cost-effective because of the use of low-cost commodity technology.

Conclusion

Large-scale projects may benefit from distributed computing, which brings together the power of several computers. More computers may be added as needed to meet increasing workload needs, making it more flexible. Although distributed computing has its drawbacks, it provides unrivaled scalability, higher overall performance, and more dependability, making it a superior alternative for enterprises dealing with increased workloads and massive data.

 

References

Hajibaba, M., & Gorgin, S. (2014). A review on modern distributed computing paradigms: Cloud computing, jungle computing, and fog computing. Journal of computing and information technology22(2), 69-84.

Jo, K., Kim, J., Kim, D., Jang, C., & Sunwoo, M. (2014). Development of autonomous car—Part I: Distributed system architecture and development process. IEEE Transactions on Industrial Electronics61(12), 7131-7140.

Jadeja, Y., & Modi, K. (2012, March). Cloud computing concepts, architecture, and challenges. In 2012 International Conference on Computing, Electronics and Electrical Technologies (ICCEET) (pp. 877-880). IEEE.

Snezhko, E. V., Kovalev, V. A., Prus, A., Dmitruk, A., & Kharuzhyk, S. (2010). DISTRIBUTED COM PUTTING IN THE NATIONAL-W IDE LUNG SCREENING AND DIAGNOSIS SYSTEM: FIRST STEPS. Materials Physics and Mechanics9, 175-184.