Software-Defined Data Centers Are Delivering On Their Promises
By Joe Masi, Enterprise Solutions Architect, CDI
Free your organization from physical hardware dependencies, procurement delays, cumbersome provisioning chores, and costly integrations. Discover how convenient it is to scale resources up or down within a new SDDC environment.
I like to read. I have been buying books since I was a kid.
The way I buy books today is light-years different from the way I did when I was younger. Back in the day, if I wanted a book, I would have to go to a book store, spend time walking the aisles, and hope that the book I wanted was in stock.
Fast forward a few years, and many of us began ordering from an online company that just sold books! Sure, it wasn’t as fast as going to the bookstore and getting the book the same day. Plus, I had to pay for shipping. But there was a 99.9 percent chance that the book I wanted was in stock!
So, I would put it in my virtual shopping cart, proceed to checkout, and then wait (im)patiently by our mailbox for it to be delivered.
A Star is Born
At this point you all know that I’m talking about Amazon. And that Amazon didn’t stop at selling only books. They saw a demand in the marketplace and capitalized on it. As a matter of fact, just last month, it was reported that there were over 500 million products for sale on their website. Many of us, myself included, even pay a subscription for Amazon Prime to enhance our customer experience.
What’s even more astounding is that despite their rapid growth, Amazon didn’t rebuild or reconfigure their data centers. Instead, they were able to scale their existing infrastructure components and occupy additional real estate when it was required.
I have been in IT for many years on both the customer and partner side of things. Throughout my career, I have noticed that the data center business operated in exactly the opposite way.
- New products, services, and other applications took months or even quarters to be brought online.
- Similar to the old way that we used to buy books, infrastructure components needed to be procured, installed, and configured in order to begin bringing value to the organization.
- A lot of time was spent (not necessarily wasted) during these periods.
What did Amazon know that just about every other company didn’t?
The Software Defined Data Center
IT transformation and data center modernization is a common initiative that IT customers have been talking about for quite some time. They want to be dynamic—to provide consumers with a streamlined way to quickly provision and deprovision IT resources. In the past, much of the time required to bring new applications online was spent analyzing and sizing the existing infrastructure to verify whether sufficient resources existed or to determine if purchasing additional components (network, server, or storage) was required.
“Wouldn’t it be great,” these early pioneers of today’s cloud-driven economy wondered, “if we could deliver infrastructure services in a streamlined and automated fashion on-demand, when end-users truly needed them? But how?”
Virtualization of servers was the initial phase in response to this question, but what about everything else?
- Increased bandwidth might require additional network ports
- Load balancers and firewall rules might need to be configured and existing rules or policies might require modification
- Increased storage might be needed to accommodate the new workload
To address these points, I want to examine implementation strategies. When it comes to hardware deployment options, there are two main approaches to implementing a software-defined data center: buy vs. build and there are pros and cons with each approach.
An organization can purchase components on the vendor’s compatible hardware list, build the infrastructure, and stand up the environment. The challenge with this approach is that once the architectural decision has been made and the equipment has been purchased, a ton of boxes will start showing up on the loading dock and the administrative team will spend the next several days, weeks, or in some cases, months putting the pieces together. A lot of cycles are spent doing the physical configuration. Because it takes longer to integrate within the environment, the organization must wait to take advantage of the solution.
Purchasing a pre-built solution frees up administrative staff by requiring less time spent building. They can focus more of their time on providing the business with a valuable solution resulting in a competitive advantage (and hopefully additional revenue). Decisions about the logical configuration are made during the procurement cycle and these parameters (in some cases) are even applied prior to the equipment shipping to the customer data centers. Think of this as a large-scale “plug-and-play” deployment model.
After the Infrastructure-as-a-Service (IaaS) platform has been configured, customers can begin their journey towards transformation. By enabling a software-defined data center, advanced provisioning processes can be implemented with automation and orchestration. Since all infrastructure components no longer have physical hardware dependencies and are defined with virtual software, processes that previously required procurement and integration are no longer a concern. It is quick and easy to scale resources up or down within the environment.
The New Normal
One of the primary goals of cloud computing is to rapidly reduce the cost of operations and the associated infrastructure components required to run a business. Most of the restrictions in the past involved data privacy and security (worth exploring in a separate conversation… maybe my next blog!); however, these factors are no longer as limiting as they once were.
Today, I am seeing SDDC steadily becoming the new normal. It is the first real competitive public cloud offering. SDDCs enable cloud pricing models with on-premise security. Since all of the components are virtualized, they are easily portable from the data center to third-party clouds which further empowers companies to gain competitive advantages and scale without procuring additional physical resources. Future enhancements are also easily consumed via updates to the software stack totally removing what was in the past a dependency on hardware capabilities.
There is much more work that needs to be done to truly deliver IaaS. When deployed, in just a few clicks, IaaS can deliver development, production, and testing environments. Since everything is virtualized, scaling the deployments, reclaiming resources, and growing the infrastructure is greatly simplified. Using commodity hardware, automation, and special sauce software, the daily care and feeding of environments is greatly simplified.
I believe that an increasing number of companies want to operate like cloud service providers. Their customers want to consume resources in a simplified manner and gain access quickly. Similar to placing products in a virtual shopping cart, proceeding to checkout, and immediately receiving the items.
Software-defined data centers can make these requirements a reality.