By Andreea Andrei, Marketing and Business Administration Executive at The Cloud Computing, SaaS and Security Awards

This article is part of an A to Z series by Cloud and SaaS Awards, continuing with S for Scalability

The ability to scale up or down without negatively affecting an organization’s fundamental operations is referred to as scalability. In other words, scalability is defined as the capacity for scale or size change without affecting the organisation’s performance and growth.

With the rise of virtualization and cloud infrastructure technologies in recent years, a new variety of access opportunities to computing resources has emerged.

The scalability of cloud computing

Cloud computing technologies provide for widespread, flexible, and low-access networked demand on a shared number of resources for information.

These technologies make it much easier to gain access to vast amounts of virtualized resources. Thus, cloud computing provides a virtualized environment for supplying resources on the Internet on demand.

There is currently a rising need in adopting this technology to solve scientific and industrial problems with large wingspan.

The most appealing aspect of switching to a cloud infrastructure is the low cost of the adopted solution, as it does not require the acquisition of specific and expensive equipment. However, one factor to consider is the need to evaluate the computational cost of migration due to performance loss caused by communications and virtualization delays.

The virtualization of cloud

The current surge in virtualization techniques has propelled the advent of cloud computing, which provides on-demand and over-the-Internet access to a set of configurable virtualized resources, and whose creation and deployment requires little effort from the user.

As a result, cloud computing has evolved into a massive repository of computational resources, providing an excellent opportunity to establish a platform for applications that require a big quantity of resources.

This cloud capacity is mostly due to the ability of resources to be elastically scaled based on application needs and user budget. This is a technology that, like any other service, focuses on providing  computational on demand. Another factor influencing application performance when deployed on cloud architecture is the ability to virtualize any resources (databases, networks, processors, etc.) and provide them as a service (AaaS / Anything as a Service).

Despite these benefits, the main issue that must be addressed is the limited performance of Internet access, which can cause delays in the execution of certain types of applications (for example, those requiring intensive calculation with quality constraints and a large amount of information transfer).

To overcome this issue, one alternative is to establish a private cloud infrastructure in order to fully reap the benefits of resource virtualization without suffering the cost of quality degradation caused by the requirement to connect to the Internet.

Cloud computing architecture

The NIST (National Institute of Standards and Technology) has made one of the clearest definitions of cloud computing:

“A technological model that enables ubiquitous, adapted, and on-demand access in a network to a set configurable computing resource sharing (for example, networks, servers, computer equipment storage, applications, and services), which can be quickly provisioned and released with a reduced management effort or minimal interaction with the service provider.”

According to this definition, cloud computing is a model for providing computer services with a primary focus on scalability. Services are elastic in the eyes of users, which means they can expand or contract rapidly and effortlessly.

This tip helps service users understand that everything works easily and quickly, resulting in a more enjoyable experience. To do this, the cloud is arranged in a layered architecture known as the service model, where SaaS (Software as a Service) sits at the top of the stack. It is the most visible level, and allows users to access cloud-hosted apps (for example: Google App).

The PaaS (Platform as a Service) is an abstraction layer or middleware that lies between SaaS and virtualized infrastructure, and its goal is to give developers with a framework and a set of APIs to create their applications (for example, Google App Engine).

Finally, IaaS (Infrastructure as a Service) provides virtualized infrastructure resources (processing, network, database, disc, etc.) as a service, in order to provide customers with an environment with scalable resources that are elastic and pay for their use.

The main properties of cloud computing include elasticity, scalability, dynamic resource supply, and payment for use based on consumption elasticity. Yet, one of the most appealing characteristics of the cloud computing migration is the flexibility in resource administration and optimization.

The ongoing desire for increasing processing capacity, not  just from the scientific and academic communities, has resulted in significant breakthroughs in computer hardware.

High performance computing (HPC) is responsible for utilizing such computer technology resources. Resources like clusters, supercomputers, or parallel computing acquire results and/or solutions to simulations that demand a large amount of computing power.

While the capacity to accomplish a lot of processing in a short period of time is the fundamental advantage of HPC, the main disadvantage is the requirement for expensive hardware to install, maintain, and operate.

Case Studies: How Cloud's Scalability Transformed Businesses

Cloud computing has transformed the way many firms manage their activities, with benefits such as cost savings, agility, efficiency, resource consolidation, and new business opportunities. Ultimately, cloud computing represents the integration of two key trends: business intelligence and IT efficiency, in which computer power is employed more efficiently through hardware configurations and highly scalable software.

OpenStack, Salesforce, and IBM

OpenStack is a system that provides Infrastructure as a service (IaaS), allowing users to run virtualized systems. The system’s modularity allows significant flexibility and scalability, being one of its key features. This allows users to deploy its components across servers in a data center in a more convenient manner.

Salesforce is the first and most well-known SaaS application. Following along, Salesforce introduced Force.com, an integrated set of application tools and services that independent and corporate IT departments may use to build any commercial application and host it on the same infrastructure that serves Salesforce CRM applications.

IBM provides cloud computing services, which provide businesses with capabilities that enable them to manage large-scale apps and databases data via the IBM cloud. The group offers consulting services to enterprises to help them integrate their infrastructure with the network.

These are instances of businesses that offer scalable resources as services to users, allowing them to increase display capacity.
The fact that cloud computing does not require an initial investment enables businesses to have a flexible use of money, low cost for irregular use, easy handling, device and location independence, rapid innovation, and, of course, scalability.

The parameters of cloud computing scalability

 

First and foremost, we must strive to reduce communication delays. If this issue is not appropriately addressed, it might affect the performance of systems with stringent time constraints, such as real-time computing. One option to mitigate the impact of this problem is to have a private cloud architecture; this way, you will avoid relying on the public cloud and therefore eliminating the cost of connecting to it.

Second, one must aim to maximize infrastructure scalability to the extent that demand for virtualized resources increases. This scalability is determined by three factors: overhead, linearity, and isolation.

  • The overhead is the processing time required by a virtual machine prior to command execution.
  • Linearity is a quantitative measure of how closely connected a set of data is.
  • Isolation ensures that in the event of a load imbalance across virtual machines, all virtual machines receive resources evenly.

Thus, the usage of virtualization technology or hypervisors has a substantial influence on the scalability of a cloud infrastructure.

As a result, selecting the correct hypervisor is critical. It is important to note that virtualization isolates the user/developer from resource orchestration details. These use resources for virtualized resource management, which leads to the fact that the more virtualized, the more overhead there is, and hence more overload and lower performance.

In conclusion, scalability is built into the cloud, which is theoretically endless. However, if your architecture is not scalable, you will not be able to take use of the infrastructure’s full scalability. Both must operate together. You will need to identify monolithic components and bottlenecks in your architecture, identify locations where on-demand provisioning is not possible in your infrastructure, and work to restructure your application to take use of scalable infrastructure and the cloud.