[notacoward]

AWS is just not very cost-effective in terms of performance per dollar, especially when it comes to storage performance (my own specialty). It only appears that they are because of the hourly billing and a human inability to compare quantities across nearly three orders of magnitude (hours vs. months) intuitively. Now that there are hundreds of vendors with hourly billing, as there have been for a while, it's easy to see how much they suck in terms of cycles, packets, or disk writes per dollar. They still have the most advanced feature set, they have by far the best public-cloud network I've used (out of nearly twenty), there are still good reasons to use them for some things, but don't go there to reduce permanent-infrastructure costs.

[Spooky23]

I just completed a project at an organization-owned datacenter where we wasted 4 months on needless BS to deploy about 12 servers.

My team's time is easily worth $500-600/hr, so we easily wasted $300k. So the fact that my internal datacenter provider can give me a VM that costs 20% of what EC2 charges or disk that is more performant at a similar cost is interesting trivia, but isn't saving money.

[sbov]

So your whole team spent 4 full time months to get 12 servers deployed? That organization sounds rather degenerate.

We colocate at a datacenter and can get cabinets pretty easily. We've done this for over 10 years now. When we aren't growing or shrinking I spend about an extra 4 hours per month because we have physical servers rather than use something like AWS.

12 servers would probably take us about an extra 6 of our person-hours to get up and running vs AWS. If we needed a new cabinet it might take a couple days, but we aren't actively working - we put in a request, and they tell us when its ready for our use. We don't sit and twiddle our thumbs while this happens, and we do it before the development side of the project is completed.

We've talked about AWS before for the redundancy and convenience but the price and the extra headache of dealing with the inconsistent performance never made sense for our use.

[lotyrin]

> That organization sounds rather degenerate.

That may be true, but it doesn't seem that uncommon.

[vidarh]

For comparison, I moved a bunch of servers into a local datacentre a few months ago. It took us ~3 days to get a rack assigned, access arranged, and a couple of days to move and set up the physical servers for a single person. Fully loaded cost of the time spent was adds up to about $4k. Total leasing costs of the equipment (some of it has already been written off, but lets assume it was all new), rack cost, and 10Mbps CIR is about $1600/month for servers totalling about 80 x 2.4GHz cores, and ~20TB storage. Of course we need to factor in some maintenance cost, but the time cost for work spent managing the actual hardware as opposed to supporting the application environment, which we'd have to do regardless of hosting, adds a few hundred a month.

Comparing EC2 costs to what sounds like a completely botched project isn't very fair, in other words. Of course there are worse alternatives than EC2 as well.

[notacoward]

"My team's time is easily worth $500-600/hr"

Clearly not.

[lobster_johnson]

Does your anecdote translate to other organizations, though?

In my own case, my company ditched AWS in favour of getting our own rack with about 10 custom servers. We have a full-time sysadmin, so nobody's time was wasted on the transition; whatever stuff the developers (who are also $500-600/hr people) were needed for during that time was valuable, because it forced us to rewrite the deployment system, which would have been required at some point anyway.

What was the "needless BS" you had to do?

[dredmorbius]

who are also $500-600/hr people

Did you sneak an extra zero in there? Even fully-realized, I'd $100-$200/hr tops in a prime market.

[lobster_johnson]

My apologies -- yes. I was actually thinking in a different currency.

[zobzu]

OVH is specialized in deployment tho, there is NO "deployment BS". You get the machines nearly instantly. It doesn't one-click scale / unscale like AWS since you need to go purchase more machines and do the deploy, but the machines themselves are available as soon as you pay them.

[samspenc]

Agreed, the features/cost tradeoff is why we're still with AWS!