Main content

Should you consider Oracle as a cloud infrastructure vendor?

Den Howlett Profile picture for user gonzodaddy December 18, 2013
Oracle is making more moves towards the cloud but does it make sense for the company to position itself head on with Amazon? I worry this might prove a diversion that spreads resources a little thin.

During yesterday's Oracle earnings call, there was plenty of mention about how the company plans to compete in the cloud space. In short - everywhere. Larry Ellison, CEO said:

Our intention is to sell our customers infrastructure as a service and the same customer a highly differentiated platform as a service will let us get better margins and highly differentiated suite of enterprise applications for the cloud.

I don't understand this as pitched on this call. Larry Dignan thinks it means:

What Oracle is really saying is that if the company doesn't compete in infrastructure as a service it's going to have trouble selling its cloud applications, which are much stickier.

From everything I have seen, Amazon is poised to own the pure play cloud infrastructure space. There are plenty of alternatives: Rackspace and Joyent immediately spring to mind with Microsoft Azure promising to be a top tier player. But Oracle? The company's more detailed intent gives some clues:

Down at the infrastructure level, we intend to be price competitive with Amazon and Microsoft Azure and Rackspace. So we intend to compete aggressively in, what I will call, commodity not being a bad word, the commodity infrastructure as a service marketplace.

If Dignan is correct then I am really struggling.

Elsewhere, I have seen a growing number of experiments where the hosting of enterprise applications takes place on Amazon. If anything, this combination provides customers with something they really want - low cost compute combined with high value apps. In one recent consulting engagement I asked the question: 'Can you compete with Amazon?' The answer was a long silence. That tells me all I need to know.

If on the other hand, Oracle believes the inevitable margin sacrifice is worth getting the long term customer lockin, then it might just make sense. This is not a slam dunk.

There is a growing feeling among buyers that the mega vendors have had it too good for too long. Brian Sommer blisters into this discussion with his 2014 predictions. I am not as polemic as Sommer but I do see heavy weather ahead for the mega vendors.

For example, a pitch that promises price parity with the market leader might sound compelling but right now, the last thing I'd be recommending is what amounts to a march towards a bundling deal with ANY single vendor. Why?

There are so many interesting things happening among smaller vendors that clearly demonstrate breakthrough value and not just incremental gain. Any thought of making a wholesale shift to cloud without clear on ramps to consuming multiple data sources and delivering operational analytics and new processes makes little sense unless there is a clear need to refresh the underlying applications.

There's a bit more to this. 

Looking at Oracle's own blurbs:

Oracle Infrastructure as a Service (Oracle IaaS) delivers Oracle Engineered Systems hardware and support for a monthly fee, with no upfront capital expenditures. Oracle IaaS combines the security and control of on-premise systems with unique features of cloud computing, including Capacity on Demand, which enables businesses to access and pay for peak CPU capacity only when needed.

I can almost hear the cloud purists in the crowd rearing up on hind legs but when you think about it Oracle's idea of IaaS is little more than a deferred purchase option for replacement data center compute capacity. Under what circumstances would I consider this when I may well already have long term deals with the likes of IBM? 

In this scenario, this only delivers billions in value to Oracle if there has been a significant turn in the renewal cycle. I have not seen data to support this theory.


I'm not convinced that mega vendors offering infrastructure is anything more than a defensive play. Even then, I'm not sure about the rationale because their business models don't suit the economics of an Amazon style operation. In short, playing win the IaaS space, however they choose to define it, is something of a distraction.

Oracle is brave. Signaling to the market it is prepared to take a margin hit in order to capture market share takes the kind of cajones we expect of this vendor. It is not a done deal.

Should you consider Oracle IaaS?  Absolutely. While we might like the look of Amazon, they're not always the correct fit. But...understand what you're likely getting into. Ellison has already told you!

A grey colored placeholder image