Two weeks ago Salesforce had an outage. Now it's outsourced to AWS

Re: the endless blame-game opportunities

Amazon, who outsources

Dude, what are you smoking?

It is the real success secret in cloud. It is called DIY and Eat Your Own DogFood. Down to the last nut and bolt in the power distribution unit of the bitbarn and last line of software. Even if it looks more expensive at one particular level it cost saves elsewhere.

That is why Google, Amazon and Facebook (with Azure catching up) have been so ridiculously successful from an infrastructure perspective. This is also why all vendor w*nkfests like HPs and Dell's miserable attempts at cloud have been miserable failures. When you outsource a blame game you also outsource the margin with it. You also now pay for the vendor ensuring that the blame is least likely. Example - do you see a vendor slotting a MB on a pizza tray without an enclosure in your bitbarn? F*** no - their own risk control will not allow it. And the cost adds up. At every level. So when Uncle Jeff comes around for your bacon you suddenly realize - your "cloud" built out of boxes which you outsourced to Quanta, Foxconn and Asus running software which you outsourced god knows where else, connected via routers and switches you bought from Cisco and built inside barns you hired and leased instead of owning (and so on) cannot compete.

Outsourcing for a customer works only at _SMALL_ scales. The moment you get into the Amazon and Co territory the economies come from in-sourcing the whole stack. And owning it - all the way to the land under it (inclusive of the f*** mineral rights).

Re: the endless blame-game opportunities

Just move from Oracle please.