(WEB HOST INDUSTRY REVIEW) — After building out huge data centers around the world, Dell has gained a considerable knowledge of how companies are using IT.
In his Tuesday HostingCon keynote, “TCO From the Cloud – What Hosters Need to Know,” Dell global marketing director Andy Rhodes presented a list of lessons Dell has learned about total cost of ownership by working with a new category of customers that have been emerging over the past few years. “They were buying and deploying devices on a scale never seen before,” Rhodes said, noting that it forced Dell to re-examine some of its ideas on IT services.
One of his main points was that each business will have a unique TCO model. “Your TCO model will depend on processes… everyone starts from different starting points,” Rhodes said. These factors may include different SLAs, labour costs, and existing IT infrastructure, as well as many others. And it gets complicated very quickly.
Given that many companies neglect a good deal of the costs – for instance, missing out on deployments because of scale should be built into the TCO model. “Get every dollar of cost in there, because it’s going to affect your bottom line if you don’t,” Rhodes said. It is also important to be flexible enough to accommodate the new infromation, and changes in the market.
Firms should also be sceptical of who is giving them advice. Given that vendors have their own interests at heart, they tend to make the model presented to customers more beneficial to the vendor.
Rhodes’ next point is that forward-looking businesses aren’t satisfied with the status quo. “There’s been a massive change in infrastructure” going from large servers to blades, and now we’ve in the age of shared infrastructure, he said. These companies have had to deal with an accelerated pace of change. Not adapting soon enough, and delays in rolling out solutions can, therefore, cost companies dearly. “An enemy of progress is the status quo.”
Rhodes’ third lesson is that the most expensive server or storage node is the one that isn’t used, or the one you don’t have when you need it most. The server at the margin is crucial. Having too few servers will make the company unable to deal with the workload, and too many, of course, result in inflated hardware costs.“If you have excess inventory, it’s bananas that are going rotten.”
Prefacing his next point with the fact that he’s a hardware and not a software guy, Rhodes recommended that firms don’t let bad code dictate the hardware architecture. Sloppy code that doesn’t taking advantage of virtualization steals resources because it necessitates “belts and suspenders” to compensate in the form of added redundancy, excess capacity, etc. And these issues can be solved by changing code – although it’s not always simple.
Rhodes recommends that company buy base elements and innovate around their strengths. “Don’t reinvent the wheel,” he said, noting that Windows Azure, Joyent, Canonical, and Rackspace’s openstack offer “cloud”stacks at the IaaS and PaaS level “that are robust and ready to deploy.”
Finally, Rhodes recommends that marketers not waste time on “cloud washing,” because the relevency of using the term “cloud” is becoming less and less. Instead, they should talk to their customers about real pain points and how to solve them. “Get away from that word, it’s not a valuable marketing commodity… customers want to know what it can do for their business.”
So, it seems that cloud best practices will help companies reduce their physical infrastructure, which seems to be a bit counter-intuitive, given that Rhodes is representing a hardware provider. But it makes sense. Given the never-ending list of projects for IT staff, and as they drive down costs, their business will grow, and they’ll be able to increase their IT spend for innovative efforts. “What we’re hopign to do is let you do more with less.”
