Thorsten von Eicken – a former academic colleague of Amazon CTO Werner Vogels and the brains behind RightScale, one of the organizations best positioned to comment on the Amazon "cloud" – has heavily criticized Vogels and company for providing so little information about the massive outage that hit their service last week and …
Am I the only one who thinks this translates as:
"Actually we haven't got a clue what happened"
and from the silence of AWS on the matter.. its either:
1) they dont know what happened yet (which is troubling) or
2) they know exactly what happened, but still considering how to slant it in such a way that they can still market their architecture as 'highly reliable' and have a soft landing after this disaster. So the name of the game is image damage control while a) this can happen again and b) other 'regions' remain susceptible to the same so called 'network event'.
You Had Me at 'EHLO'
Like any of the other great failover disasters, this problem was exacerbated by the recovery plan. Load control is more important than anything, if you lose that everything else falls down around your ears. Happens with Email Storm failovers, happens with distributed hosting failovers, happens with NAS replication failovers.
So one of the grrrrrreat promises of The Cloud® is that The Cloud™ is soooo reliable and you don't have to worry about all that annoying DR / BC stuff anymore. That's nice. But then you have outages such as this, and then I see that Amazon has a 99.95% uptime SLA. That's not good enough, and it's certainly not good when Amazon clearly doesn't want to be clear about what/how/why this happened. They're being quite cloudy about it really. And I don't want to single out Amazon - Google has had its problems as well, and as more stuff gets clouderised™, I'd expect that such services will become more and more attractive to ne'er-do-wells and hax0rs etc.
I shall heed the amazingly prescient words of that great prophet, Mick Jagger, and stay off of their clouds.
99.95% uptime is 4 hrs 23 mins down time a year.
And it s a *service*
As in the service *provider* handles things, not you.
Part of "handle" being to keep customers (they people who are paying for this) informed in a way they can plan what to do next.
Is anyone seeing parallels with RSA?
May I remind all CEO's...
If it's not on your infrastructure it's no longer your data.
I think you mean...
If it's not on your infrastructure it's no longer your datum. Singular, you see.
Off you run now, go play....
@John G Imirie
That's patent bollox and you know it.
In the computing world at least, outsourcing is the oldest profession. In many circumstances it provides systems that are more reliable, more secure, and better performing than in house systems. Sometimes it can even save money, but that is not always the primary goal.
How it is implemented, both technologically and in terms of commercial agreements, is what makes the difference between successful and failed outsourcing. But that is equally true of in house systems.
Outsourcing is the oldest profession
That explains why one partner in the deal seams to be getting screwed by the other.
If I don't get payed this month because the outsourcing company handling the pay role screwed up It's not them I'll be asking compensation from.
Here's what really happened at amazon over the weekend http://www.youtube.com/watch?v=m3wrBFuGK2A