Pentagon admits it's now probing conflicts of interest at AWS over $10bn JEDI cloud deal

Re: Oracle...

I'm sure Oracle have never had any revolving doors with any government either. Kettle meet pot.

Re: Oracle...

There has always been some revolving doors between the government and industry. DoD is well known for this as ex-military work for military contractors and contractors get hired by DoD. Nothing new really. The only issue is how close the person was to specific project and whether they could influence the specs to favor a specific vendor too much. All government specs will favor a group of vendors by either by specific technical requirements or by the lack of some requirements. This is the way the game is played at all levels. You try to get some requirements in or out that favor a group of vendors over another. If done right, there will be a group (3 to 5 ideally) of qualified bidders so the bid is not a sole-source. Sole-sourcing is not easily done as it requires a lot more paperwork to justify it in Feraldom and the paper pushers hate extra work.

So whether this will have any effect, I doubt it unless the specs are so specific that only AWS can meet them. Then there might an issue but the key is what is in the paper trail.

Kool-Aid

"That's the reason that all the main cloud providers have multiple regions and multiple availability zones, so that a failure in one area can be worked around. Nothing new here, multiple sites that are physically separate and logically linked has been the norm for a very long time even for on-prem data centres.

Going multi-cloud means that you can't take advantage of the specifics of any cloud vendor, but have to settle for the lowest common denominator functionality, which kind of defeats the purpose of having access to the vendors new sparkly technologies and the benefits it can bring."

But only half of the story, isn't it?

"the specifics of any cloud vendor" are frequently nothing more than incremental improvements—if they are even that—primarily intended to make it difficult and expensive for a client to move elsewhere. The idea that this or that cloud supplier has some super-secret-sauce which a client simply must have is just laughable. The services provided by the top three cloud vendors do not vary by performance, security, reliability or even cost sufficiently to make a telling difference. Choosing one is like selecting a mobile phone provider—cutting through all the contrived "advantages" and "USPs" and deliberately, arbitrarily complicated pricing mechanisms to discover that the latter are designed primarily to make it difficult for potential customers to perform a simple like-for-like cost:benefit analysis.

"sites that are physically separate and logically linked" sounds great but (a) any such advantage is replicated by having different providers, and (b) if the probelm, as is so often the case, is one of software, there's a fair chance it is affecting geographically diverse centres anyway.

The short answer is that if you lazily, short-sightedly, greedily allow your multi-million or -billion enteprise to become too dependent on any third party, whether outsourcer or cloud, they will contrive to entrap you and bleed you dry.

So, if you've got a scrap of sense: Don't.