I want to go on record with my preliminary investigation of Oracle licenses obtained through traditional sources and those provided by Amazon. There are two different philosophies at work here. Oracle and Amazon are at odds. Amazon will win. Either Oracle prices will be allowed to get cheaper, or people will move off to other products. There is already a move heavily under foot. Call it big data, call it horizontal scaling, call it by any other name and it will still take an architectural design to get there. As easy as it gets, as much as it becomes a commodity, it requires a certain amount of skill, knowledge, confidence… and luck.
If you are planning on using Oracle at Amazon there is a lot to learn about licensing.
The bring-your-own version is not practical. If you are still considering Amazon, is the edition available adequate? Do you need options? There are currently approximately 38 extra cost options. These are in addition to the base-entry price of the database itself.
Chances are, if you “need” Oracle for any reason other than you already have it, you should probably consider the move to some other architecture as you move to the cloud.
Now, I am all for administrators knowing how to cluster data on Amazon; but, why are you considering optimization? If you’re just doing it because it’s a feature, it’s there, and you’re using it. Okay. However, if you are seriously trying to optimize Oracle, you probably ought to be considering Oracle (formerly, Sun) hardware.
Face it. The Oracle database was made for Sun hardware. Their chips are valued, by Oracle, at twice the value of an Intel or AMD chip. Hence the processor calculation in computing license cost by using <point>5 to arrive at number of CPUs.
And I assure you, there is Oracle hardware and software built for Oracle so high up the totem-pole that I can never expect to see it, much less have experience with or be in charge of…. ever.
Some sort of cost analysis will be looked at in the future as I gather actual numbers from running instances. I will also be experimenting with some of the options, looking at other configurations, and considering tools.
My guess is that Oracle will allow Amazon prices to continue, whatever they are. Oracle, as a stand-alone product will not get any cheaper. Oracle as a company will get bigger. They’ve had a good leg up; but Amazon will “win” if it’s ever considered a race. The philosophies are too radically different to have any other result.
Someone may tell me that Oracle has their own cloud solution, and yes, I’ve heard some of the terms and requirements. I also remember Larry Ellison saying that “Oracle is the Operating System”, and what a laugh we got off that! Yet, a lot of what he has predicted over the years has worked out.
Does every cloud has a silver lining? I am open to alternatives, particularly from companies that press the issue: like Amazon. Oracle, I love you. Amazon, I plan to love you for years. Welcome!