Feasibility Proposal on Cloud Computing

Differentiate between cloud computing models and their uses

The best advice when making the decision to switch to a public model is choosing a model based on the needs of the company. This decision enables convenient network access to a shared pool of configurable computing services. Cloud Computing is storing data or accessing programs via the internet instead of using a hard drive (Raza, 2020). Many organizations are switching from their own private servers to contracting the service from reputable providers, to be relieved of the responsibility of maintaining the software and guaranteeing security. To begin with there are three common cloud service models which provide different levels of control, flexibility, and management. Software as a Service (SaaS) vendors host the service via the internet. In this case the vendor handles all of the maintenance. Platform as a Service (PaaS) provides developers a platform to create customized softwares and applications. This choice requires the provider to manage the servers, storage, and networking, leaving the responsibility of managing the application to the developers. Finally, Infrastructure as a Service (IaaS) offers computing resources via the internet and takes care of the servers, storage, and networking hardware (Kumar, 2022).

These shared resources over the internet have different deployment models that define the services being consumed and the responsibility model for the provider. To expand, the deployment model expresses the cloud architecture, the scalability of resources, and the level of customizability. The public cloud does not involve the user owing any of the hardware or the infrastructure, though the resources provided are open to the public. The benefit of this option is the elastic scalability, resource efficiency, and of course relieving the stress of high capital investment (LaunchDarkly, 2022). Private Cloud is best used when handling sensitive data and when greater control is needed over data residency, stricter data access, and data privacy measures. The option of a hybrid cloud combines public and private cloud features, meaning the organization is using the public cloud but also handles their own on-premise system. Finally community cloud allows few organizations within the same “community” or market to utilize this cloud.

There are many factors that are involved when making the division of which deployment model or which service model. For those organizations with limited resources should consider SaaS because they can save money instead of expending efforts to design and develop the software themselves. For larger companies with enough resources should consider IaaS because they are given complete control with highly customizable technology that can be altered to fit the goals. The best option depends on how much money, time, and effort needs to be saved (Rosencrance, 2021). Considering the deployment options, switching to a public cloud may be a better option to save cost if the data is not sensitive. While the private cloud may be a great option for large organizations that handle sensitive data and have the necessary funds to maintain the hardware. While hybrid may be a great option for those companies in the process of cloud migration. A community cloud is usually used by banks because the resources are tailored for that specific community and its needs.

Benefits and Drawbacks of the Cloud

Many organizations are adopting the cloud delivery models instead of on premise models for many reasons that benefit their business. The biggest benefit is being able to access the application at any time from any place via the internet. Also the initial cost to move is lower than the capital expense of creating an on-premise model. The operating cost of utilizing a cloud software depends on the resources used, the system also scales to meet these demands. While the benefits of an on-premise model includes the level of control the organization has over the data being stored, the hardware used, and the software being maintained.

The biggest disadvantage of using the cloud is that in the case of weak internet access or during downtime. Downtime is a tech industry word for the duration during which the IT system is unavailable or offline. Another drawback of using a cloud software is the long-term cost, it ends up surpassing the initial set up expense of an on premise model. It is also great to mention the decline in customization allowed when using the cloud. Depending on the service model a cloud can be altered to fit the needs of an organization but typically the application is configurable to an extent. But depending on the complexity of the development, it may not be able to handle it. A disadvantage of an on-premise model is the responsibility to maintain the system. In the case of a disaster, the organization is at fault and must remedy it. They must manage the server hardware, the software, initiate data backups often, and facilitate storage space. Another drawback of the on-premise option is the time it takes to complete installations or updates on the servers to each device, one by one (Cloud vs on premise software, 2017).

Cloud Deployment Models

Some benefits of a private model is the users control over the service, the policies, and the IT operations. This also relates to the benefit of customization, developers can alter the platform to their own needs. Another benefit of the private cloud is the data security; it is easier to manage who has authorized access to the information. If these benefits are not a good enough reason to switch then perhaps the public cloud might be a better fit. It requires minimal investment, which is great for smaller organizations. With the low set up cost, users only pay for what they use, which relates to the next benefit of dynamic scalability. Paying for what you need also means you have on-demand resources for any project. Finally the public cloud also relieves the responsibility of infrastructure management and all other maintenance (Sameekshakhandelwal1712, 2022).

With all of the benefits mentioned about the private cloud, there are a few drawbacks. This includes the cost of maintaining the entire cloud, companies are responsible for the cost associated with housing the hardware and also the actual software supporting the cloud. This also relates to the cost of choosing this model, there will need to be money spent and effort used. Choosing a public cloud also comes with its own set of disadvantages, especially including the fact that there is no guarantee of data security. Meaning your data could be compromised, to no fault of your own. Another con to mention is the low customizability of public clouds. If employees are not trained and given the proper resources, it can be a tough time to troubleshoot problems that arise.

Considerations of Cloud Computing

Organizational issues are any difficulties employees encounter that could prevent or affect them from fulfilling their responsibilities. This should be considered when changes are made in effort to help operations because it could backfire or come with disadvantages. These operational drawbacks and organizational issues should be considered before deployment. Perhaps employees are not knowledgeable of the new system, this could be an organizational issue of moving to the cloud. Another disadvantage could be managing the system and integrating the cloud because every platform is different. Organizations may not have the proper controls or knowledge initially to monitor performance of the platform (Taylor, 2021).

Technical issues are the problems associated with the software or the hardware of a system. An example of this would be issues related to integrating the new cloud because of incompatible processes or authentication errors. Another display of technical problems with cloud computing is interoperability and portability. Challenges of Interoperability occur when organizations attempt to move data across multiple cloud ecosystems. There may also be problems encountered when handling data encryption during the migration process (Anand, 2023). Oftentimes, the problems with switching to a cloud from an on-premise model are exacerbated when switching between cloud servers. It can be difficult for organizations to finalize the transfer of data to a better cloud provider, this is called vendor lock-in. (Problems with cloud computing, 2022).

Big Data vs. Structured Data

Big data is a combination of structured, semi, and unstructured data that can be mined for information (Botelho, 2022). Big Data is composed of data in large columns with high velocity and wide variety. Collecting big data can be done by operational systems to automate the process from multiple sources. For example, ETL data pipelines involve extracting data from different sources, transforming it to a clean formatted state, and loading it into a data warehouse. Data preprocessing is a method of filtering, transforming, and encoding data. This step is done to get an overview of the data to catch missing datasets, identify outliers, or remove inconsistencies. Since big data is too large and complex for traditional methods, there are large-scale processing platforms like MapReduce to process datasets automatically. Apache Spark performs faster computing tasks with features like fault tolerance, failure recovery, and data partitioning (Gaikwad, 2018). Structured data is organized data that is formatted before being processed and can be collected manually. It is usually stored in data warehouses such as relational databases. There are hands-on methods to complete the preprocessing step. Techniques such as data sampling and stratified sampling involve grouping to manipulate and examine a representative subset of the original subset (Gupta, 2022).

Volume, Variety, and Velocity of Big Data

Big data comes with its own set of challenges, as it is difficult to guarantee data quality, data storage, and integrating data from different sources. It's important to recognize the three V’s of big data; Volume, Velocity, and Variety. Due to the large volume of data (and even the variety of data types) it is far too time consuming to process through traditional systems, such as SQL. This means there is a capacity on how much, what type, and how fast data can be collected, processed, and analyzed before automation must take place. Many times the appeal to utilize big data leads companies unprepared for the large amount of effort it takes to make significant insights. Since big data holds potential to reveal competitive advantages, companies invest efforts in collecting and housing large amounts of information. However, it is only part of the battle, the return on investment happens after analyzing and applying the insights gained from the patterns found. There is such a thing as too much data, a sign of this could be when the efforts to wrangle data get in the way of exploring hypotheses and reaching conclusions to make smart decisions. Too much data, often referred to as analysis paralysis, happens when too much “noise” slows down the decision making process (Yiu, 2019).

References

Anand, B. (2023, January 18). Top 15 cloud computing challenges [with solution].

KnowledgeHut. Retrieved March 24, 2023, from https://www.knowledgehut.com/blog/cloud-computing/cloud-computing-challenges

Botelho, B. (2022, January 5). What is Big Data and why is it important? Data Management.

Retrieved March 24, 2023, from https://www.techtarget.com/searchdatamanagement/definition/big-data

Cloud vs on premise software: Which is best for your business? Xperience. (2017, February

24). Retrieved March 24, 2023, from https://www.xperience-group.com/news-item/cloud-vs-on-premise-software/

Gaikwad, M. (2018, May 24). Preprocessing of big data. Excellarate. Retrieved March 24,

2023, from https://www.excellarate.com/blogs/preprocessing-of-big-data/

Gupta, S. (2022). Hands-on methods for structured data pre-processing. enjoyalgorithms.

Retrieved March 24, 2023, from https://www.enjoyalgorithms.com/blog/structured-data-pre-processing-hands-on

Kumar, A. (2022, August 8). Top 3 cloud computing service models: SAAS: Paas: Iaas. Cloud

Training Program. Retrieved March 24, 2023, from https://k21academy.com/amazon-web-services/aws-solutions-architect/cloud-service-models/

LaunchDarkly. (2022, October 4). Cloud deployment models: Explaining and comparing the

5 main models - launchdarkly. LaunchDarkly. Retrieved March 24, 2023, from https://launchdarkly.com/blog/cloud-deployment-models-explaining-and-comparing-the/

Problems with cloud computing. FutureLearn. (2022, October 25). Retrieved March 24, 2023,

from https://www.futurelearn.com/info/courses/key-topics-in-digital-transformation/0/steps/262275

Raza, M. (2020, August 31). Public vs private vs hybrid: Cloud differences explained. BMC

Blogs. Retrieved March 24, 2023, from https://www.bmc.com/blogs/public-private-hybrid-cloud/#:~:text=Or%2C%20perhaps%20you%20use%20the,private%20cloud%20for%20sensitive%20data

Rosencrance, L. (2021, December 20). SAAS vs. iaas vs. paas: Differences, Pros, cons and

examples. WhatIs.com. Retrieved March 24, 2023, from https://www.techtarget.com/whatis/SaaS-IaaS-PaaS-Comparing-Cloud-Service-Models

sameekshakhandelwal1712. (2022, November 29). Cloud deployment models.

GeeksforGeeks. Retrieved March 24, 2023, from https://www.geeksforgeeks.org/cloud-deployment-models/

Taylor, G. (2021, January 13). 6 organizational challenges for Cloud Services. CloudTweaks.

Retrieved March 24, 2023, from https://cloudtweaks.com/2021/01/6-organizational-challenges-for-cloud-services/

Yiu, T. (2019, November 26). How much analysis is too much? Medium. Retrieved March 24,

2023, from https://towardsdatascience.com/how-much-analysis-is-too-much-e1dfc5b37cbb#:~:text=This%20is%20commonly%20known%20as,down%20the%20decision%20making%20process