Final Research Paper

docx

School

SSAT Degree College *

*We aren’t endorsed by this school

Course

Subject

Information Systems

Date

Nov 24, 2024

Type

docx

Pages

Uploaded by CaptainFalconPerson690

CS63170G222 Final Research Report A study of DB Transaction Processing / Coordination used in a Cloud environment DharaniDhar Kotlapati Advanced Database Campbellsville University

Abstract The phrase "cloud computing" is one of the most contemporary and fashionable words to be used in the field of information technology. The concept of centralizing data storage and processing in an environment with scattered data led to the development of a methodology that was later given the name "cloud computing," which has been developed to represent this strategy. Its short- and long-term objectives are to give cloud customers flexible and on-demand package alternatives, allowing them to have significantly greater flexibility, speed, and reliability all at once. This will be accomplished by using a straightforward "utility computing paradigm." It argues that it will be able to offer pricing based on demand, a decrease in expenditures connected with information technology, as well as the capability to swiftly scale down or up IT infrastructure (Mara et al., 2017). Transaction processing systems that, among many other things, operate in multi-processor and cloud-based contexts are a primary focus of this study. Every significant company that offers cloud computing services has settled on a configuration that is unique in comparison to those of its competitors. As a consequence of their efforts, the author of this article will examine a number of them, such as the basic techniques that were developed as a consequence of their work to improve Cloud Transactions. Keywords: Cloud Computing, Information Technology, Processing System, Transaction

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help

Table of Contents Abstract .................................................................................................................................... iii Chapters .................................................................................................................................... 1 Introduction .............................................................................................................................. 1 Problem statement .................................................................................................................. 2 Relevance and Significance .................................................................................................... 2 Research questions ................................................................................................................. 3 Barriers and Issues .................................................................................................................. 3 Literature Review ..................................................................................................................... 4 Approach/Methodology ........................................................................................................... 8 Amazon Elastic Compute Cloud (EC2) ................................................................................. 8 Findings, Analysis, Synthesis ................................................................................................. 10 Conclusion ............................................................................................................................... 13 References ............................................................................................................................... 14 v

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help

1 Chapter 1 Introduction One of the most significant developments in software technology over the course of the last half-century has been the capacity to rapidly process data. Applications that handle transactions are vital to the functioning of governments, corporations, the financial system, the transportation sector, and the military. According to Maria et al. (2017), such programmers are employed in a number of operations, including order processing, finance, digital reservation processing, telephone switching, and other jobs of a similar kind. Large corporations that deal in computer hardware & software, like IBM, Amazon, Oracle, as well as Dell, use many transaction-processing systems in their business operations. These companies generate annual revenue in the hundreds of billions dollars due to the transaction processing items and services they provide to their customers. The amount of money that they would get as payment for these products or services would be in the hundreds of billions of dollars range. This kind of computer service is provided to consumers over the internet, where they have the ability to see software regardless of the fact that the underlying programs and services are kept on a variety of servers located in various locations across the globe. Cloud computing allows us to shift the burden of data processing from widely scattered, locally- hosted devices like laptops and desktop PCs to centralized, powerful data centers. For instance, any device could be capable to solve certain complicated equations by simply sending the appropriate inputs to an data center service which is able of answering in a minimal amount of time if done out in such a manner. In this manner, the process may be carried out. On the other hand, worries over the security of data & app usage have dramatically developed over the course of the last several years.

2 Problem statement In this article, the author will analyze a number of these ways, with a particular emphasis on the underlying principles that each approach employs in order to improve the efficiency of an cloud transactions (Awad et al., 2021). Transaction processing programs that, among many other things, run in multi-processor and cloud environments are the primary focus of this investigation. When it comes to a cloud services they supply, every significant cloud provider does have an architecture that is better to that of their rivals. A few of the most important firms in the area of cloud computing include Google, Amazon, Yahoo, and Microsoft. Some hardware manufacturers, such as IBM, HP, Dell, as well as Intel, are also among the most important corporations in this industry. Relevance and Significance In this article, the author conducts an analysis of the benefits that customers receive from purchasing a product packaged in a container. This provides customers with the opportunity to select the packaging option that caters to their requirements the most effectively by taking into a account the particulars of their unique circumstance. The offer of (practically) endless scalability is, without a doubt, one of the most important benefits that come with cloud computing. One important advantage is that IT administrators no longer need to fret about how to cope with heavy traffic. The availability of these capabilities via cloud computing may result in increased adaptability and dependability in the usage of both hardware and software as well as their administration. Significant savings are realised in both "production time" and "manufacturing expenses" as a consequence. Research questions 1. What aspects of an transaction processing have led to the field's widespread acceptance as a reliable software technology?

3 2. What precisely is meant by the phrase "cloud computing," as well as how does this notion compare and contrast with the idea of an computer service that is provided over the internet? Barriers and Issues Numerous studies have been conducted to assess the security, scalability, and efficiency of cloud computing systems because of their meteoric rise in popularity. Among the findings is that there is a notable performance gap between Hadoop, a free and open- source Java-based application, and more conventional (SQL-based) database systems in certain scenarios (Awad et al., 2021). Recent research has analyzed the tradeoff between cloud computing's low prices and its lack of consistency while handling transaction processing activities.

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help

4 Chapter 2 Literature Review NoSQL databases are preferable when it comes to the processing and storing of vast amounts of data because of their ability to handle a broad range of characteristics such volume, diversity, and velocity. This gives them an advantage over traditional relational databases. These kinds of databases are used in a wide range of user applications that need a considerable amount of data that is consistently highly available and easy to access on an regular basis. These kinds of databases also meet the requirements of the apps' users. Despite this, they do not actually enforce or demand high standards of an data consistency, and as a result, transactions are not made any simpler. In addition, they do not make any progress toward the goal of simplifying business processes. In this article, we will look at consistency- aware apps that are hosted on MongoDB and Riak. These two databases are classic examples of NoSQL databases that store documents and key-value pairs, respectively. This group is responsible not only for the development of transaction methods for NoSQL databases but also for the investigation of the effect that transactions have on the data integrity and productivity of applications that are built by end users (Mara et al., 2017). A benchmark known as YCSB + T, which is developed from the Yahoo! Cloud Services Benchmark, is used in order to do an evaluation on the suggested approaches. The findings of the tests suggest that it is feasible to utilise the supplied methods to establish strong consistency in MongoDB in addition to Riak without severely hurting the performance of either database. This is the conclusion drawn from the findings of the studies. Experiments are also carried out to determine the amount of consistency that is attainable by using the transactional systems MongoDB and Riak. The word "Metamorphosis" has been given to the topic that has been discussed in this article by those working in the field of online business

5 (MetaQ). This communications system contains features, which allow for a higher degree of detail than is otherwise possible. There is a large variety of software that may be used to store message offsets in a variety of locations (Jianwei et al., 2017). One of the characteristics that makes it an ideal option for usage in application contexts is the capability to support a huge number of transactions in addition to real-time needs. This is only one of its many advantageous qualities. Even more importantly, a distinct version of a MetaQ implementation that is currently being utilised by both Taobao and Alipay is known as RocketMQ. This version of the system was recently implemented. This is a change that will have a significant impact in the future. The nature of MetaQ has been shown to perform well in both typical online transaction apps as well as large-scale Internet apps of a same kind, as shown by real-world applications. In recent years, there has been a significant increase in research into cloud computing, particularly study into data replication methods and the uses of such systems. If the amount of copies is increased as well as when they are dispersed over various locations, it will ultimately result in an increase in the total cost. This study suggests two bio-inspired methods as a method of enhancing the selection and placement of any data copies in a cloud environment. These are 2 components of the process which are grossly insufficient at the current moment, thus this research recommends improving them using bio-inspired algorithms. Utilizing the first strategy that was discussed, which is known as MO-PSO, it is possible to get a data copy that has the best selection. This would be a practical option. This version is derived from the data copy which is used the most often. On the other hand, the MO-ACO strategy is employed in order to acquire the best data replica placement. This placement is decided by the shortest distance between replicas in addition to accessibility of a replicas. The MO-ACO approach is the second preferable option. In order to get the best possible position for the data replicas, this strategy is used (Awad et al., 2021).

6 CloudSim was used to do a simulation of the proposed procedure, and the results painted a positive picture of the approach. Every data centre (DC) is made up of hosts, and on each of these hosts are installed a certain number of an virtual machines (VMs). The order in which data is replicated is chosen by picking one instance at random from a pool of one thousand cloudlets. This determines the sequence in which data is duplicated. In accordance with the strategy that has been proposed, all of a replication files have been dispersed in a way that is completely random. In furthermore, in contrast to other approaches, MOACO provides a larger data availability at lower costs and with lower bandwidth use than its rivals. Having the ability to manage industrial equipment as in cloud in a flexible manner becomes an increasingly vital talent to possess. Production within the context of the Fourteenth Industrial Revolution (Industrie 4.0). A cloud control method that enables the sharing and control of integrated manufacturing capabilities is needed in order to effectively execute the idea of offering manufacturing services via a cloud environment. This is necessary in order to ensure the concept's successful execution (Adamson, Wang & Moore, 2019). In this article, an product-oriented approach is used to describe an cloud-based control mechanism that has the qualities that are essential to carry out control. It has been conceptualized as a prototype system as well as put into operation; it uses the previously stated control mechanism; and it is capable of both local and distant operation. The purpose of this article is to demonstrate how the control strategy may be used by presenting an example of a real-world application scenario. The cloud computing environment produces a massive amount of data each and every second, and the entire volume is consistently increasing over the course of time. The use of something that is known as a multi-cloud platform is something that cloud service providers use in order to manage the data and satisfy the requirements of their customers. Service providers could be able to boost the accessibility of an entity's resources by collaborating on the design of dynamic

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help

7 operations that are capable of being carried out simultaneously in an environment that has many clouds. The service providers are able to take advantage of these processes. In this article, we take a look at a data storage system that goes by the name DA-ICP (Dynamic Level Based Integrity Checking Protocol). It was designed with a multicloud setting in mind from the beginning. By employing the Provable Data Possession (PDP) methodology, a customer that outsources data to an untrusted multi-cloud could benefit from the peace of mind of knowing that the server in question does in fact own the original data. Because of this, the user is able to verify that the server really has the data without having to download the data (Anwarbasha, Sasi Kumar & Dhanasekaran, 2021). Picking a random group of blocks from a server may provide a probabilistic proof of ownership, drastically reducing the cost of establishing such a proof. It has been shown that the efficient as well as protected data that's been outsourced was encoded with the aid of Efficient-PDP and that it was decrypted with the help of public key cryptography (EPDP). The DA-ICP that was provided at every stage of the research project demonstrated an accuracy level that was the greatest possible at 96.78 percent. When applied to DA-ICP, the proposed method makes advantage of multi-cloud, which results in an output which is more effective than that generated by other ways that are presently in use.

8 Chapter 3 Approach/Methodology This analysis was compiled after an exhaustive search for related literature on transaction processing was conducted in the CU library and on the Google Scholar search engine using the phrases "transaction processing," "apps of a transaction processing," & "Information Technology Architectures in an Distributed Environment." The outcomes of these studies encouraged more study to be conducted utilising a variety of databases. These databases include, but are not limited to, science direct, Web of Science, and IEEE Access, amongst others. We did get rid of any study that had been carried out more than once so that we could make the procedure more streamlined and get rid of a research that was not required. In addition, for the sake of this study, any research that was carried out in a language other than English was discarded. Amazon Elastic Compute Cloud (EC2) Amazon's Elastic Compute Cloud, often known by its acronym EC2, is generally recognised as being among the most successful web-based services which Amazon.com makes available to the general public. EC2 is sometimes shortened to just EC2. Amazon Web Services (EC2) is an web-based service that delivers scalable and also secure computing capability in environments that are housed in the cloud. In the meanwhile, it is planned to make web-scale cloud computing increasingly accessible to the general public, although at the same time it was developed with programmers as the primary audience in mind. The term "Microsoft Azure" refers to an cloud computing service which is provided by Microsoft.

9 The Microsoft Azure cloud platform is a collection of cloud-based services that are built on top of the.NET Framework and SQL Server. The platform was created by Microsoft. The concept known as "Platform as a Service," that is more of an umbrella term, is the primary focal point of this service (PaaS). Both the entire application frameworks, like as Google AppEngine, and the hardware virtual machine instances solutions, such as Amazon EC2, are in direct rivalry with one another. This level of competition is good for the industry as a whole. Users on the receiving end of the Windows Azure service have the ability to execute applications and store data on servers controlled by Microsoft and made available over the internet. The service makes this accomplishment feasible. The user mode is the only one that allows access to software that is operating on Microsoft Azure; administrators do not have permission to access the program. Microsoft Azure is the company's brand name for the cloud computing service which it offers to its customers. The.NET Framework and SQL Server serve as the foundation for the Microsoft Azure cloud platform, which is comprised of a collection of cloud-based service offerings. This service places a significant emphasis on something that is often known as "Platform as a Service," that is an all-encompassing term (PaaS). Both the entire application frameworks, like as Google AppEngine, and the hardware virtual machine instances solutions, such as Amazon EC2, are in direct rivalry with one another. The industry would benefit from this healthy competition. Using the Windows Azure service, end-users were able to execute programs as well as store data on computers that are owned by the Microsoft as well as are accessible through the internet. This is made possible by the Windows Azure platform. This is a result of the service that was provided. Access to applications running on Microsoft Azure is restricted to users; administrators cannot get in.

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help

10 Chapter 4 Findings, Analysis, Synthesis As a consequence of the widespread use of cloud computing, a substantial amount of investigation has been carried out to evaluate the dependability, adaptability, and quality of cloud computing infrastructures. It is a term that is used to describe a certain "class" of software packages that have the capability of enabling transaction-oriented applications to be executed via the Internet. Online transaction processing is what is meant by the abbreviation OLTP. OLTP is an abbreviation that you could also see used. The classic "distributed transaction" idea (Mara et al., 2017) is improved upon by the distributed transaction concept, which eliminates the need for many nodes and databases to complete key transactions. This eliminates the substantial performance difficulties that plague enterprises using the old "distributed transaction" concept. The second difficulty arises in circumstances in which businesses have a widespread geographical reach and are needed to connect with business associates who could be located in every region of the world, such as the case with Amazon. In these kinds of situations, it can be difficult for businesses to maintain a consistent level of communication with their business partners. When compared to other approaches, the old paradigm of "distributed transactions" is notoriously sluggish and unreliable for the reasons listed above (Awad et al., 2021). The Information Technology Architectures of an Globally Distributed Environment The area of cloud computing has unexpectedly become a home for distributed database architectures, more often referred to as DBAs. The standard design for a multi-tier database & application will be discussed first, then two more variations of this architecture

11 will be broken down and explained. In addition to partitioning, replication is a component of these versions. Classics Depending on the specifics of the requests, such requests are sent from clients to a computer that can act as both a web server as well as an application server at the same time, and that computer is both accessible and providing both of those services concurrently. Following the receipt of an HTTP request from an user, a web server will execute the request by carrying out the instructions included within the code using a programming language that might or might not include embedded SQL. A database server receives the outcome of this processing, at which point it processes the request, replies to the user, as well as updates the database. When data is sent between a database server and a database, the data is transferred in actual chunks of information (64K blocks), and the GET and PUT requests serve as the interface between the two. This takes place throughout the process of sending data from one device to another. Partitions The following points best illustrate the distinction between partitioning and traditional approaches to database design: The database literature has many distinct instances of partitioning methods in their various forms. These strategies include, but are not limited to, round-robin vs hashing, range partitioning, vertical partition versus horizontal partition, as well as round against hashing, to name just a few instances each. The website known as Force.com, which is in charge of hosting the application known as "Salesforce," was the first to construct a database design that was compatible with an cloud technology and permitted partitioning. A complete server-side application stack is what the partitioning of Force.com is

12 made up of. This stack consists of a variety of components, one of which is web application servers (Awad et al., 2021). Replication Replication functions in a way that is similar to that of partitioning when it calls for the employment of many database servers, which each stores an exact duplicate of the whole database. Maintaining consistency is indeed the single most critical part of any process that involves replication, since it is the factor that decides whether or not all copies will be correct. In order to realize the objective of making a replication totally undetectable, the first occurrence of each query is transmitted straight to the Master server. A non-transparent replica means that only certain apps will send queries to the database server in the charge of a Full copy to ensure that the data is always up to date, with any notifications about committed changes coming from the Master server. Methodology, measurement, and even the actual carrying out of the plan are all extremely important aspects to take into account. The purpose of this performance study was to investigate the scalability of various cloud service providers with regard to bandwidth, along with the expenses that are involved with making use of such services. The following objectives might be attained with sufficient effort: This was accomplished by applying a TPC-W benchmark to a alternative services that were outlined in Chapter 4, and also assessing WIPS as well as cost, which ultimately resulted in a change in the EBs. In addition to this, this was carried out in order to accomplish this objective (i.e., number of simulated concurrent users). As was said before, in the interest of this study, we raised the load from 1 EB, which is equivalent to a moderate amount of work, to 9000 EBs, which is equivalent to a substantial amount of work (Jianwei et al., 2017).

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help

13 Chapter 5 Conclusion Within the realm of information and communication technology, the concept of "computing in the cloud" is one that is both novel and exciting. A system that is based on the Internet and in which applications and data are stored on a variety of different servers is what is meant by the term "cloud computing" (locations). Customers now have access to cloud computing services from a very small number of companies. These companies include Microsoft, Twitter, Google, as well as Oracle. CloudTran and Gnubila are 2 of the greatest cloud networking solutions that are now accessible because to their user-friendliness, stability, reliability, and flexibility in relation to cloud services like Google App-Engine and Amazon EC2. In addition to this, they are capable to do all of the tasks that were mentioned before in order to make use of a straightforward utility computing paradigm. In this study, we analysed the advantages that come with buying each product as part of an package. This gives customers the flexibility to choose the option that best satisfies their requirements in the most effective way possible (Awad et al., 2021). Recommendation As part of its mission, this project seeks to improve fault tolerance, dynamically provide the right amount of an load balancing to respond to incoming traffic, and detect unhealthy conditions.

14 References María Teresa González-Aparicio, Adewole Ogunyadeka, Muhammad Younas, Javier Tuya, & Rubén Casado. (2017). Transaction processing in consistency-aware user’s applications deployed on NoSQL databases. Human-Centric Computing and Information Sciences , 7 (1), 1–18. https://doi.org/10.1186/s13673-017-0088-3 Awad, A., Salem, R., Abdelkader, H., & Salam, M. A. (2021). A Novel Intelligent Approach for Dynamic Data Replication in Cloud Environment. IEEE Access, Access, IEEE , 9 , 40240–40254. https://doi.org/10.1109/ACCESS.2021.3064917 Adamson, G., Wang, L., & Moore, P. (2019). Feature-based function block control framework for manufacturing equipment in cloud environments. International Journal of Production Research , 57 (12), 3954–3974. https://doi.org/10.1080/00207543.2018.1542178 Anwarbasha, H., Sasi Kumar, S., & Dhanasekaran, D. (2021). An efficient and secure protocol for checking remote data integrity in multi-cloud environment. Scientific Reports , 11 (1), 1–8. https://doi.org/10.1038/s41598-021-93073-3 Jianwei Liao, Xiaodan Zhuang, Renyi Fan, & Xiaoning Peng. (2017). Toward a General Distributed Messaging Framework for Online Transaction Processing Applications. IEEE Access , 5 , 18166–18178. https://doi.org/10.1109/ACCESS.2017.2717930

Final Research Paper

Related Documents