Welcome!

Thoughts on enterprise IT

Dustin Amrhein

Subscribe to Dustin Amrhein: eMailAlertsEmail Alerts
Get Dustin Amrhein via: homepageHomepage mobileMobile rssRSS facebookFacebook twitterTwitter linkedinLinkedIn


Blog Post

There's More Than One Entry Point for Cloud and Virtualization

Match your enterprise's entry point to pain points

If you are someone who works in the cloud computing and advanced virtualization sector day in and day out, you may have a perspective that makes it easy to forget that this movement is still very much in its infancy. Personally, from time-to-time, I find it easy to forget how early these technological shifts are. For me, that lapse is usually momentary, owing to a particular question I hear repeatedly from enterprises: "Where do we start with cloud and virtualization?"

A significant number of enterprises are in the early stages of formulating a strategy and execution plan to leverage the benefits of cloud computing and virtualization. An obvious part of that execution plan is establishing an entry point that is both consumable and valuable for the enterprise. For that reason, when enterprises ask me where to start, I turn it back around to them and ask, "Where are you biggest pain points in service delivery and management?"

I make it a point to ask this question because it seems as if many players in the industry imply that the only starting point is at the infrastructure level. In other words, you should look for solutions that allow you to virtualize, manage, and govern compute infrastructure like servers, storage, and networking (whether this infrastructure is on-premise or off-premise is irrelevant to this discussion). This is a fine starting point... for some situations.

I suspect one can attribute the industry's inclination to push users towards an infrastructure entry point to a few different factors:

1) Both on-premise and off-premise solutions are visible and well known (think VMware, Amazon EC2, IBM PowerVM, etc.)

2) This level typically requires little change. Essentially, it is a faster, more consistent, and more efficient way to provision the same services an enterprise already uses. The difference is the services live within some sort of virtualized container.

3) Under-utilization of servers is a widespread problem. Reports consistently put average server utilization between 7-12%. Virtualization equips users to easily drive up utilization.

While these may be good reasons for some to start at the infrastructure level, it certainly does not mean that an enterprise should only consider starting at this level. Consider the following scenario (one that is not at all far-fetched based on enterprises I talk to):

The biggest pain point for Enterprise A is in maintaining a highly performant production environment. The environments are fairly static, meaning that once deployed, the services tend to stay up and running for a long period. Enterprise A does not deploy new services into these environments often, and they average a reasonable server utilization rate north of 50% (comparatively this would be more than reasonable).

If Enterprise A were interested in cloud computing and looking for recommendations of where to start the journey, I certainly would not suggest the infrastructure level as a means to solve their most pressing pain point as stated above. Their server utilization is respectable, and while faster deployment times may be nice, the fact that not many deployments are taking place means that this will not necessarily be a huge return on whatever the necessary investment may be.

If I were to suggest a starting point for Enterprise A on their cloud and virtualization journey, I would be inclined to point them toward solutions that provide a means to virtualize, manage, and govern applications (or ones that enable this for both infrastructure and applications from a single control plane) instead of the underlying infrastructure. Enterprise A stands to gain much more from a system that provides autonomic, policy-based management of application instances and application resources to ensure that their applications perform in accordance with the needs of the business.

I hope that focusing on such a narrowly-scoped scenario does not dilute my point in this post. I simply mean to say that if you and your enterprise are considering embarking on a cloud computing and/or virtualization journey, start by examining the pain points in your IT organization (especially around service delivery and management). Once you identify the pain points, prioritize them to understand which ones are the biggest inhibitors. Once you have that list, look to solutions that provide you the most bang for your buck. Oh, and remember, buck means acquisition and implementation costs. Do not get fooled by the price tag alone!

More Stories By Dustin Amrhein

Dustin Amrhein joined IBM as a member of the development team for WebSphere Application Server. While in that position, he worked on the development of Web services infrastructure and Web services programming models. In his current role, Dustin is a technical specialist for cloud, mobile, and data grid technology in IBM's WebSphere portfolio. He blogs at http://dustinamrhein.ulitzer.com. You can follow him on Twitter at http://twitter.com/damrhein.