Cloud Computing Essential Characteristics

This chapter takes a look at all the essential characteristics of Cloud Computing, including: elasticity, scalability, resource pooling, flexible pricing models, service layers, and ubiquitous network access.

Elasticity & Scalability

Elasticity refers to the ability of the infrastructure resources to grow or shrink in a dynamic manner (Complete Controller, 2018). That is, the growth or shrinking happens as needed – when the workload changes. This paves the way for maximum utilization of resources. Furthermore, it makes it possible for resources to be used cost-effectively. However, it is important to note that not all are able to enjoy the benefits of elastic resources. If you operate in environments that do not undergo sudden demand changes, the elasticity feature of Cloud Computing is not something that you will benefit from. It may actually be a negative trait where certain applications require guaranteed performance (Complete Controller, 2018).

An example that needs Cloud elasticity is retail which undergoes seasonal changes. For instance, Black Friday is a period when people tend to spend huge on products due to presumed discounts and enticing offers. During such times, the system could experience sudden increases in demand. Rather than using huge amounts of money to buy additional infrastructure to provide needed support for a short period of time, an elastic solution finds its place here. The additional infrastructure is only paid for on a pay-as-you-go basis and can shrink when demand reduces. This goes a long way in helping you save costs.

Cloud scalability

Cloud scalability is all about the ability to increase the size of workload within current infrastructure without reducing performance. Also, scalability entails expanding with additional infrastructure resources without difficulties. You can view scalability from both a vertical and horizontal perspective. When viewed vertically, scalability would involve scaling up within a system. Horizontally, you could scale out multiple systems although not linearly (Bastide, 2019).

Scalability could be seen in virtual desktop infrastructure (VDI). The employee population has a direct impact on the expected number of desktops. In order to support the maximum number of users, the total purchased services should be enough for supporting all users logged in at a go without degrading performance.

On-Demand Self-Service

Cloud Computing is known to provide resources on demand. That is, resources are provided to consumers whenever they need them. This is a feature that is possible due to automation and self-service. Automation entails things happening in the Cloud without user intervention. Self-service entails the user doing all that they need to acquire the service at an individual level without the intervention of the IT department. When the user exploits self-service to make a request, the Cloud infrastructure automatically processes it from the provider’s side without any human intervention.

On-demand self-service requires a great input when it comes to planning. For example, a user may wake up any day and request a new virtual machine. The user may want the machine to begin functioning optimally within a few minutes. If the underlying hardware takes a long time to relay this information to the provider, the user would end up being disappointed. Thus, it is vital that the provider keeps tracking how resources are being used so that they can plan for the future.

Still on on-demand self-service, the user interface has a great implication on this Cloud Computing characteristic. Simple user interfaces are the best approach. The provider can never assume that the user has technical IT knowledge or that they have a detailed IT setting within the organization.

Multi-tenancy & Resource pooling

Resource pooling entails sharing various computational abilities. This has the advantage of making sure resources are properly utilized. With resource pooling, the organization ends up requiring fewer resources, therefore saving on costs.

The characteristic of Cloud Computing whereby a program can serve several consumers (tenants) while remaining isolated from the others is known as multi-tenancy. With the aid of multi-tenancy models, a Cloud service provider is able to meet the needs of multiple consumers. As resource pooling makes it possible put together large-scale IT resources for use by various tenants, multi-tenancy ensures that the various tenants are effectively served through assigning and reassigning of resources.

Resource pooling allows enterprises to save costs while giving the latitude for providers to flexibly maneuver through the Cloud Computing world. This concept is based on the notion that tenants will not continuously need to use the resources available to them. Instead of allowing a given resource to be available to a consumer who is not using it at a given time, they can be made available to another consumer who is in urgent need. Therefore, providers can serve many customers than they would have done, had the resources been dedicated to a single tenant.

Economics & Flexible Pricing Models

The IaaS and PaaS service models for Cloud Computing have a variety of pricing models given that the services have been greatly commoditized. Software as a Service up to this moment remains to have a flat price for a certain period of time (Suliman & Ammar, 2015).

There are a number of factors that control demand and revenues in Cloud Computing. Among these factors, pricing takes the highest precedence. With the aid of pricing, Cloud providers are able to determine: (1) the degree to which service provisioning ought to be done for consumers, and (2) how pricing relates to other facets like grant discounts and provisioning.

Generally speaking, there are two pricing models used for this technology, namely: fixed pricing model and dynamic pricing model (Belo, 2014).

  • Fixed pricing model – The price charging in this model does not undergo changes and the Cloud provider determines price based on the type of resource being assessed in advance. For instance, Amazon AWS charges a rate of $0.15/GB(Maitland, 2010). Fixed pricing model is considered the simplest to understand given that it is pretty straight- forward. However, the model is regarded as being unfair to all customers since each customer has unique needs.
  • Dynamic pricing model –This is a pricing model which changes based on market trends. The pricing mechanism in place is used to calculate the service price for various requests. Even though you do not get a straight-forward amount to be paid, the model has the element of being fair in terms of charges(Maitland, 2010).

Abstraction & Services Layers

An abstraction refers to an individual, high-level concept. It is the simplest concept possible. For instance, when one writes a report, there is always an executive summary which helps readers to avoid having to wade through the whole report. Essentially, the executive summary is the abstraction – no wonder scientific journals refer to them as an abstract.

Applied to Cloud Computing this concept is very important. When combined with encapsulation, it helps prevent things from degrading. Encapsulation refers to taking an object and hiding it from view. In so doing, the user is prevented from having to worry about complexity of single pieces. This combination of abstraction and encapsulation is very powerful. The two constructs have made it possible for users to access provision applications in Cloud with much ease.

Besides abstraction, service layers are also an important characteristic of Cloud Computing. The Cloud is made of several layers in a similar manner that an onion has layers. There are three major layers of the Cloud, namely: Software as a Service – SaaS, Platform as a Service, and Infrastructure as a Service – IaaS. SaaS is the top layer while IaaS is the foundation layer. You will realize that graphic representations make SaaS appear to be the smallest layer. That assumption is far from the truth. SaaS is based on both IaaS and PaaS. It is the most scalable, affordable, and can be easily maintained.

Ubiquitous Network Access

Ubiquitous networking is also referred to as pervasive networking. It entails making sure that there is constant connectivity through the distribution of wireless technologies and communications infrastructure in the environment (Schouten, 2014).

Cloud services are based on the Internet infrastructure; therefore they offer an ever-present availability as long as users have Internet connectivity. A Canadian CEO is able to go about his duties while on a business trip in the U.S., given that he has access to the company’s resources that are hosted on a server in Singapore.

Ubiquitous network access is all about the ability of Cloud service to be accessible to any person from any location as long as they have required credentials and Internet connection (Schouten, 2014). To ensure that there is omnipresent network access for numerous Cloud services, it may be necessary for different security technologies, interfaces, transport protocols, and devices to be supported (Rouse, Ubiquitous networking, 2017). This is a kind of access that is available only after the Cloud service architecture is tailored to meet specific needs of various Cloud service consumers.

Leave a Reply

Your email address will not be published. Required fields are marked *