Live data from Hacker News

The cloud has failed.

struct3.com

1–7 of 7 posts

Re: The cloud has failed.

#2
"when a thing that cannot possibly go wrong goes wrong it usually turns out to be impossible to get at and repair."

When your server fails, you know how to fix it. When the cloud fails, you hope that somebody will be able to fix it soon. When p2p fails you're probably out of luck for a long time.

Re: The cloud has failed.

#3
post #2

"when a thing that cannot possibly go wrong goes wrong it usually turns out to be impossible to get at and repair." When your server fails, you know how to fix it. When the cloud fails, you hope that somebody will be able to fix it soon. When p2p fails you're probably out of luck for a long time.

People are always citing p2p and bit torrent as the magic answer to stuff. But p2p requires mutual self-interest.

if you've ever been on a torrent with no seeds and a hundred leeches you'll know what I mean.

Re: The cloud has failed.

#5
The actual title of the article is "Has the cloud failed?", can we please not put our own editorial spin on things when submitting articles? We'll read them and make up our own minds.

Re: The cloud has failed.

#6
post #2

"when a thing that cannot possibly go wrong goes wrong it usually turns out to be impossible to get at and repair." When your server fails, you know how to fix it. When the cloud fails, you hope that somebody will be able to fix it soon. When p2p fails you're probably out of luck for a long time.

When your server fails, you know how to fix it. When the cloud fails, you hope that somebody will be able to fix it soon.

Well, not really. If your application is in a datacentre somewhere, whether that's a hosting provider or somewhere within your organization, then it might be a physical box, or a VM, or a running in a cluster or a cloud, it doesn't matter. What does matter is that there are capable infrastructure specialists on-hand to deal with any issues. That hasn't changed and won't ever change.

Re: The cloud has failed.

#7
post #2

"when a thing that cannot possibly go wrong goes wrong it usually turns out to be impossible to get at and repair." When your server fails, you know how to fix it. When the cloud fails, you hope that somebody will be able to fix it soon. When p2p fails you're probably out of luck for a long time.

I guess you can use the skype outage of last year (I think??) as an example of this, as skype is still somewhat p2p. And it wasn't a short outage, not compared to EC2 etc.