Guy using a tablet sitting on a cloud

Why Scale Matters in a Cloud Strategy

One of the most important characteristics of cloud computing is that it’s scalable. The ability to expand and contract infrastructure resources on demand is what makes the cloud so powerful.

Being tied to the physical constraints of hard-drive space, CPU, memory and bandwidth is one of the limitations of physical hosting. But with cloud computing, you can focus on building your infrastructure for success rather than worrying about whether your in-house infrastructure can handle the constraints of success in the first place.

The National Institute of Standards and Technology (NIST) defines cloud computing as “…a model for enabling ubiquitous, convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, servers, storage, applications, and services) that can be rapidly provisioned and released with minimal management effort or service provider interaction.”

NIST goes on to describe the five essential characteristics of the cloud:

  • On-demand self-service
  • Broad network access
  • Resource pooling
  • Rapid elasticity
  • Measured service

While not explicitly defined as “scalability,” there are two key phrases here that connote this term, specifically computing resources that can be “rapidly provisioned and released” and “rapid elasticity.”

When something is elastic, it can expand and then snap back to its original size. With infrastructure, that means various components can be added or removed from the infrastructure to make it larger or smaller, as needed. From an infrastructure standpoint, the cloud allows you to programmatically add or remove servers using a control panel or an Application Program Interface (API). This is where the “rapid provisioning” comes into play.

Controlling the Components of Scale in the Cloud

Depending on the cloud environment, you may be able to get much more granular with your control. While the most typical example of scaling involves simply adding cloud servers to or removing cloud servers from your infrastructure, you may also be able to scale at a “component” level; in other words, you may be able to add or remove RAM, hard drive space, CPUs and even networking.

There are specific use cases for needing to scale these individual components. Operating systems and applications are often very RAM-hungry, meaning they will consume all available memory and consequently run more efficiently when more RAM is available.

Database servers tend to perform better with more RAM and CPU cores. And since many web, social and financial services generate terabytes of data, hard drive space — potentially in the form of cloud or attached storage — is a core element of scalability. A Big Data implementation, for example, may actually require numerous hard drives configured in various RAID arrays in order to store data and make it highly available. If your data storage is physically constrained, your business cannot scale.

Where Cloud’s Ability to Scale Counts Most

Why would you want to scale your infrastructure in the first place? Here are a few specific industry-use cases:

  • E-Commerce Seasonality: If you run an e-commerce site, business will ebb and flow. During the holiday shopping season, you may need to have extra processing power to handle an increase of shoppers. Then, during the off-season, you may need to scale back on the infrastructure in order to avoid overspending and having that infrastructure sit idle.
  • Data Analysis: If number crunching or statistical analysis is your business model, you may need to run minimal infrastructure when you’re not processing data sets, and then add more processing and/or storage capability when it comes time to analyze the data.
  • Social Media: If you are managing or developing social media applications, you will need infrastructure and storage that scales. Many social media applications are driven by Big Data, which means that a multitude of data is created with each social interaction. From a scalability standpoint, you will most likely be scaling up (adding more processing capabilities) and out (adding more servers) to handle growth.

Don’t Wait to Plan for Scale

All businesses, regardless of size, should consider their scaling strategy. If they don’t, they are setting themselves up for failure. Although it’s fairly obvious why large enterprises need to be elastic in their infrastructure provisioning as they add new products or services, smaller businesses need to be agile as well, especially if their business models are tied to seasonality or important product announcements that can drive traffic to their sites.

There is nothing worse for a business than to have its website or web hosting service perform lethargically or time out. This represents a loss of revenue and may alienate customers. If customers are turned away at the “digital door,” they may not come back. If your infrastructure is programmatically configured to monitor traffic spikes, or if you have analyzed your traffic or infrastructure utilization over time, you can get ahead of the curve so that you have infrastructure ready and available when it’s demanded, scaling back when the need (and traffic) subsides.

Moving to a cloud hosting environment means that you can grow your business and online presence easily and stay agile in the process, allowing you to save not only on infrastructure cost and management. Ensuring your infrastructure scales as your business does isn’t a nice-to-have — it’s a must-have of modern IT.

[image: BsWei/iStock/ThinkStockPhotos]

ABOUT THIS CONTRIBUTOR
Michael Sheehan is a technology expert with a passion and expertise for cloud computing, enterprise IT and gadgets.
Back to top