Everyone is talking about going to the Cloud. Some reports by Industry analysts predict that all customers will be in the Cloud in the next 5 years. One of the primary drivers moving everyone to adopt the Cloud is cost. In theory, running workloads in the Cloud will be cheaper than buying the infrastructure to run on-premise. If done correctly, this is a true experience. However, it can actually cost 2-5 times more if there are calculation errors. These errors throw a monkey wrench in the whole enterprise, and leave a bad taste in the mouths of those who feel there is a total misconception about the advantages of Cloud computing. In effort to avoid the common pitfalls of Cloud computing, which are known to obliterate cost savings, there are a few key factors to keep in mind on the journey to the Cloud.
One primary misconception about the Cloud is that not all workloads are meant to go into the Cloud. The Cloud was developed for elastic capabilities in compute, storage and network. Simply put, organizations pay only for what they need, and only when they need it. The Cloud is perfect for unpredictable workloads, such as those in retail, which are known to be busy during the holiday season. The Cloud may not be ideal for organizations that have a constant ERP or processing data need, with constant predictability and no downturn in the amount of work. This is an example where using the Cloud may cost more and not be suitable for the Cloud since these types of workloads are always on. The Cloud pricing model is most advantageous for workloads that experience a lot of fluctuation in load, from huge system overuse to almost stagnation. To use an old adage, the Cloud pricing model is cost-prohibitive for workloads that are so predictable you could set your clock by them.
Because not all workloads are meant for the Cloud, and not all Clouds are meant to hold all workloads, it’s important to understand the differences and know the Cloud’s strengths and weakness. Infrastructure-as-a-Service (IaaS) can be run on any Cloud for x86, but each Cloud is designed for a specific workload. For example, Oracle Cloud is the only Cloud that has the capability to run Solaris in the Cloud, just like IBM Softlayer is the only Cloud that can support AIX. While you can run Linux in Azure, Microsoft SQL is a better fit for Azure. Choosing the right Cloud for the workload allows you to leverage existing licenses and spend wisely where you’ve already invested.
Virtual server sprawl is a given in the on-premise data center. Lines-of-Business asked for servers left and right, when they found out that servers could be provisioned without having to wait for hardware purchases. Along with those dynamics, trailed the capacity issue of running too many servers. The saving grace for understaffed, data center managers, was that you had to stop deploying when you ran out of resources. However, in the Cloud, resources are always there. You will only stop deploying when the monthly bill exceeds the available funds. The issue becomes when end users, DBAs, and anyone in the IT department of an organization can access AWS and Azure workloads, and spin up new demands for the Cloud implementation on a daily basis. Without proper planning, controls, auditing and procedures, these Cloud resources can be easily forgotten. Meanwhile, the costs are adding up for the organization.
The journey to the Cloud is an undertaking that needs to be carefully planned. Seek outside resources who can offer unbiased, vendor agnostic opinions, as well as several alternatives for you to consider. Realize you may have to spend more time and money upfront in order to gain long-term cost savings. Understand the different workloads in your organization and assess their capacity over time before unilaterally deciding to move everything to the Cloud. To use another old adage, be careful not to fall into the trap of Monkey See Monkey Do.
For more information, please visit www.pcm.com/cloud
About the Author
Field CTO – Director of Datacenter Transformation Services
JOIN THE CONVERSATION
Share your thoughts and questions in the comment section below. To get the latest news from PCM, follow @PCM on Twitter, join us on Facebook, or connect with us on LinkedIn. To get the latest news sent straight to your inbox, join our newsletter.