Live data from Hacker News

Effectively using AWS Reserved Instances

stripe.com

121–123 of 123 posts

Re: Effectively using AWS Reserved Instances

#121
post #96

Earlier quoted context omitted.

No they didn't. Because you are a hosting company. Not all companies have your expertise - most don't.

The hyperbolic difficulty that folks are applying to running your own hardware is laughable. It is as if by miracle alone that civilization made it through the precarious transitionary period where we racked our own servers into the Wonderous Utopia of Cloud. I definitely wouldn't tie my own shoe laces, only the largest orgs need laces, almost everyone can focus on walking if they use Velcro.

It’s about staff and focus.

Right now we have 3 sysadmins for our own DC, and they can’t keep up with all the maintenance (upgrades on infra are the worst. Software quality on switches/SANs/appliances is terrifying. And those are better than quality of Server management, BIOS, etc. Even VMware has caused lots of pain with patches and upgrades in recent years.

If you don’t ever patch/upgrade to fix all the security vulnerabilities that exist at the infrastructure layer maybe “do it yourself” is cheaper. But that doesn’t fly in the banking industry.

We’re half-in the cloud and moving more there just to keep our head above water. We’re spending way less on cloud services than the $500K/year it would cost to double our infrastructure staff and keep it all in house.

Re: Effectively using AWS Reserved Instances

#122
post #96

Earlier quoted context omitted.

The hyperbolic difficulty that folks are applying to running your own hardware is laughable. It is as if by miracle alone that civilization made it through the precarious transitionary period where we racked our own servers into the Wonderous Utopia of Cloud. I definitely wouldn't tie my own shoe laces, only the largest orgs need laces, almost everyone can focus on walking if they use Velcro.

It’s about staff and focus. Right now we have 3 sysadmins for our own DC, and they can’t keep up with all the maintenance (upgrades on infra are the worst. Software quality on switches/SANs/appliances is terrifying. And those are better than quality of Server management, BIOS, etc. Even VMware has caused lots of pain with patches and upgrades in recent years. If you don’t ever patch/upgrade to fix all the security vu…

I've written a lot on this thread. I think the linchpin of this entire thing is how much salaries for technical people have lifted.

If you can get a qualified dev/dc engineer for 50 or 60K maybe you rack your own. But when facebook is paying all-in median wages of 240K/year things start to look different. I understand not everyone is going to be able to work at fb and not everyone lives in California. But when one of your critical inputs doubles or triples in price, of course that's going to adjust how things get done an in industry.

Re: Effectively using AWS Reserved Instances

#123
post #76
post #31

Earlier quoted context omitted.

Why not just rent dedicated bare metal from a large provider?

I'd expect that to have nearly all the disadvantages of cloud (other than virtualization). One is still locked into the provider's pricing structure. One is still locked into the provider's ISP choices. One is still locked into the provider's internal network architecture choices. One is still locked into the provider's limited hardware choices. The last one is the biggest one, if only because there are so many oppor…

Most providers I have seen that rent hardware offer unlimited customization when it come to the box you are renting. Network architecture would obviously be off the table though.
Post reply on HN