Maybe I'm reading this article wrong, but it strikes me that Scoble is unaware that lots of sizable startups don't own servers. What's the difference in this case between using ec2 instances and renting servers from a provider like LayeredTech, SoftLayer, ServerBeach, Rackspace, etc.? I understand there's a huge difference if you're bringing up -- and especially down -- ec2 instances in real time as demand changes, but beyond that, it's renting a server for $70 a month. So, for by his definition, Wordpress.com is a server-less company with 300+ servers in three colos.
We were considering using EC2 & S3 for our upcoming launch, but we did some calculations, and found the minimum cost for 24/7 service comes to about $70 per month not including data or bandwidth charges. Once a site has greater usage and needs to scale, it seems hard to beat, but for the early days it doesn't seem to make financial sense.
Basically, any storage on the machine itself could go poof from one minute to the next - or at least you have to think of it that way, because once it goes down, it's gone.
Therefore, you need to have some kind of distributed storage architecture, and automated recovery for said system.
For extra points, do the whole thing with no external server with a stable IP.
If you are not willing to risk 70$ per month you are in the wrong business my friend, or you are woefully too cash-strapped even the smallest bootstrap should be able to afford that.
And what's the alternative going down the Twitter road? I think not...
Always have a reasonable scalability plan preplanned, it doesn't have to be fully implemented but you should devote some thought cycles to the problem and have a road map for scaling.
Um, who are you to discourage this person because he doesn't want to blow $70 a month needlessly rather than spend $20 on a decent server slice? Not trying to troll here, but this comment makes you sound like a pompous ass who's never bootstrapped a company.
Sure, it's important to have a scalability strategy planned, but that's nowhere near the same thing as wasting money on excess capacity before you need it. I suppose a few years ago you would have advised the HotOrNot guys that if they couldn't afford to maintain a standing account with Akamai before launching, they were in the wrong business too.
I wasn't trying to be condescending and I apologize if I hurt any one. I have actually bootstrapped quite a few startups (crashed and burned horribly on the first 3, only the 4th one actually made me a bit of money). And when I thought I didn't have enough money saved for what I thought I needed as bare minimum I suffered through another couple of months in my day job before taking the plunge.
The entire advantage of Amazons services is that you don't need to pre-purchase too much capacity and you can scale relatively painlessly so long as you designed your server architecture for it.
I admire the HotOrNot guys for their "beg, burrow, steal" approach, I wish I had even a fraction of it. And that was exactly the right thing to do back then. But now the infrastructure has improved so much that you don't need to do it for servers.
Keep your energies for the other hundreds of battles you'll need to fight on your way & good luck
Not everyone has a stable day job before they get into a startup. It's not in everyone's best interest to suffer through one for a few years first either.
I thought it was the opposite -- it's cheaper than buying your own servers and co-locating to start, but like any rental, eventually you hit the inflection point and it costs more.
Also, multiply by 4 or 8 depending on what instance size you are using; so it's $72, $288, or $576 a month, per box.
If you're considering "cloud computing", you should also check out Joyent. (http://www.joyent.com/) They're different than Amazon EC2, though... rather than just being able to scale up by adding additional virtual servers, your servers are also "burstable" up to the maximum physical capabilities of the physical machines. Pretty cool stuff. I just got an Accelerator for my in-progress Facebook-powered site, so we'll see how it turns out. (By the way, Facebook applications get free entry-level Accelerators...)
I'm interested in hearing more about this, too. It sounds like the two offerings combined (service plus storage) make for a complete hosted solution. Is this correct? And how are domain services handled? By a third party?
EC2 on its own is pretty comparable to a "standard" hosting setup.
This is their "small instance" setup.
1.7 GB of memory, 1 EC2 Compute Unit (1 virtual core with 1 EC2 Compute Unit), 160 GB of instance storage, 32-bit platform
The usual thought is to use dynamic dns to point to a server you have setup in an EC2 instance, so yeah a third party.
However, this isn't really an all or nothing proposition. You can certainly plug and play different components in to complement your existing setup and still save money / increase reliability.
If you have a solid setup, but are exceeding bandwidth limits you can move static files to S3.
Want better testing rigs, you can make EC2 instances for each and just turn them off when you're done.
EC2 and S3 are great tools, but you still have to do the homework to make them fit your needs.
Has anyone here done this for their web app? I am considering some grid-server space from MT, but if EC2 is good, it seems like it's cheap; is this a viable option?
Stay away from MT - the grid is slow even for simple apps, and they go down on a regular basis. Have also had bad experience with Dreamhost.
I recently signed up w/Joyent and am happy with it so far, have also heard good things about SoftLayer, ServerBeach, RackSpace, Layered Tech...some of the other names you hear thrown around on here.
EC2 is probably a better option once you've been around for awhile and that economy of scale starts working in your favor.
