We applied to TechStars in NYC first because we were based in NYC and we got in right before the early app deadline. After meeting David Tisch at TechStars 4 A Day he flat out told us he doesn't understand our space so it would be hard for him to pick us, because part of his decision is to determine how he as the program director can help accelerate our growth.
We did become a finalist but weren't selected and he recommended us to TechStars Boulder. So we flew out there for TechStars 4 A Day and went through the process again.
Full post: http://www.quora.com/Startups/If-youre-rejected-from-an-incu...
Stored backups (for free) of servers that you can set up at a later date instantly via an api
It's not about being a $5 vps for me, it's about being able to pay $0.014 when I need a small vps for 2 hours.
Our service is mostly CPU bound, embarrassingly parallel, and so far, 90% of usage is centered on a time slot that's about 8 hours per week long.
We will save a ton of money by scaling down during the off hours, which means only Amazon, Rackspace, and a few others make any sense at all.
I lurked for less than a day on Lowendtalk and got an idea of the most respected/high quality hosts such as:
Prometeus (and IperWeb)
I've had a box with most of these (apart from Ramnode which is highly regarded) and i can say i've had an awesome experience.
I think "sketchy" is a good characterization when a pick of the "best" hosts have severe problems.
As someone who have used quite a few servers from LEB over the years, there is no way I am going to put anything serious in most of their offering. There are some exceptions, like Ramnode, BuyVM, but DO pretty much made the choice easy for me. Why bother taking a risk?
1) Cheap price and good offering and service.
2) Now I know they are going to stick around for a while.
Remember Slicehost? They might stick around but not in the form that you like especially given that DO are playing the startup game rather than growing organically, building a solid business, and then taking money at a later stage for targeted expansion, like Atlassian and Github.
If Apple got hacked because their dev center was powered by a $10 control panel script then yeah... they'd be extremely sketchy too.
For example, I run several servers on rackspace. Their least expensive option (512MB RAM, 1 core, 20GB disk) is $16.00 per month.
DigitalOcean's is $5.00 per month. On the face of it, they are identical offerings but triple the price.
Rackspace does offer other features (load balancing, cloudfiles, and other useful things which integrate nicely with their servers.) Is that the value proposition of AWS/Rackspace over these other companies, which only give you vanilla servers?
Some estimate AWS's margins are up around 80% 
Competition, low cost hardware, building their own stack, open source etc. mean better pricing for us users
This is not to run down DigitalOcean, but if you take reserved instances into account then they don't necessarily beat EC2 pricing. Or am I missing something?
To those of you asking what is so special about DO compared to ExyExtraVps.is running stock WHMCS or whatever over at LEB it is that DO are providing the type of features, flexibility and support that Rackspace, AWS and Azure provide but at prices that are close to what the low-end VPS types offer.
If you imagine a Gartner-style quadrant for hosting with price on one axis and then features/support/flexibility on another most providers today sit somewhere along a very straight line - you either have a bunch of features and flexibility and are expensive or you are very rigid and cheap. DO is right up in high features but low price, and it seems so obvious in hindsight (Linode were almost there).
I now have Vagrant setup with Digital Ocean as a provider so setting up a box is as easy as running 'vagrant up --provider=digital_ocean' and then waiting a minute. There are tons of other tools already built up around their API and the ecosystem is thriving. I have another script setup ready to fire up additional instances of web apps.
I'm moving everything I can and that fits from AWS and recommending DO to clients also where it would fit (staging servers, dev servers, backup servers, etc.) There are still some features they need - like load balancing, multiple IP's etc. but this funding means all of that is going to be built sooner.
There is a reason why they are one of the fastest growing hosts ever. These guys are destined to get really really big.
There is also a great effect wherever digital ocean is mentioned online you find a comment thread with dozens of users saying how awesome they are. You can't buy that type of love.
 Not to mention that all of that off-the-shelf VPS software running these sites is an absolute security nightmare. The developers rely on hiding their poor code behind ioncube encoding and this has been exposed recently with some big name hacks.
 Netcraft: The meteoric rise of Digital Ocean: http://news.netcraft.com/archives/2013/06/13/the-meteoric-ri...
edit: forgot to include this earlier, but by way of a disclaimer I introduced Digital Ocean to Crunchfund but don't have a financial stake.
I understand that you can have a Vagrant provider for VMWare and VirtualBox, but what does it mean for you to have a DigitalOcean provider for Vagrant?
Also, maybe someone can drop a few words in on the typical usage scenario for Vagrant? Other than automation, how else does Vagrant help a developer, and how does it tie into the various VPS services (such as AWS, DO, etc...)?
Instead of spinning up a VM on your local machine it spins up a new droplet in your Digital Ocean account (droplet is DO's term for a virtual server).
> Also, maybe someone can drop a few words in on the typical usage scenario for Vagrant? Other than automation, how else does Vagrant help a developer, and how does it tie into the various VPS services (such as AWS, DO, etc...)?
Vagrant lets you define a server environment that you can turn on with just a simple "vagrant up". For apps that have server dependencies (who doesn't?) like databases or MQ servers, you can isolate them into a standard VM. If someone else is going to work on your app then they can get started immeidately via "vagrant up". No need to manually setup the server dependencies, it'll all automagically build itself.
Once you start using it you'll never go back to manually setting up server dependencies for dev. It pays for itself in time savings the first time you use it.
Vagrant integration with AWS or DO is simply to have the server spin up there rather than locally. Even cooler is the expermental Docker integration (as a provider) as it lets you run many Vagrant VMs with minimal overhead.
> Instead of spinning up a VM on your local machine it spins up a new droplet in your Digital Ocean account (droplet is DO's term for a virtual server).
>Vagrant integration with AWS or DO is simply to have the server spin up there rather than locally.
Every droplet costs money, why would you want to create new droplets for development, when you can spin up a VM locally? What are the advantages of spinning up remote server instances over local ones?
In my workflow I have two boxes in my Vagrant profile, one for development (local VBox VM) and one for production (DO droplet). Using Chef for provisioning, I can configure and test provisioning locally, then roll it out in production with minimal effort.
Its a really nice workflow! Although all this Docker business lately looks like it might make for an even better one.
once the server is up, it is exactly the same as if you did it by hand
Droplets only cost money when they're instantiated/on. At the end of the day, you nuke the server and can deploy it again in the morning if you need to.
This is nice as it allows you to do performance testing as well as stability and provisioning tests and your normal unit\functional\whatever tests.
While you wouldn't want to do it for every commit (unless you had your own infrastructure with space(though you could still do a partial test and profile)) it would be nice to test every merge to master.
And this kind of growth almost always means founders are looking for an early exit. Yes it's all a happy ending for the founders but what will happen to the end users?
Hardware prices don't drop so much at this scale unless you're buying in extremely large quantities. Even then and they don't drop as much as bandwidth (except for out of lease equipment and I doubt DO does this).
VPS in general is insanely profitable, if you know what you are doing.
The employee aspect will always be there, DO can still make money by not taking outrageous profit margin like AWS and Rackspace.
You don't really charge a client for the bandwidth costs incurred during a DDoS if you choose to absorb the attack. You don't charge a client when a RAID controller fails and you need to restore xTBs of data (xx man hours). Oh and you need to purchase backup nodes, which you're not getting any immediate return on.
You need an office. You need to factor in travel costs. Overtime pay. Health insurance, payroll costs, etc. Ads...
You can buy a 1U server that could fit 100 DO sized droplets for about $5,000. Scale that up a little bit, and you can see that it might be profitable although low margin.
Equally, at our company, we're paying DO for many different machines at $160. I'd love to know their ARPU number, I expect it is actually significantly above $5.
From just two days ago, here on HN:
"Digitalocean.com has misconfigured their network in a way that allows for anyone to monitor customer network traffic." https://news.ycombinator.com/item?id=6157747
- Their website is easy to use and clear
- Access via a simple API
I've used them for some side projects - although they aren't feature-full for bigger ones, they are great to use.
DigitalOcean are only special because they "play" the start-up game - lots of hype, marketing, get journalists on their side to write nice things about them, etc. - and this gives investors something they can bite into.
The difference between DigitalOcean and other low end providers is that DigitalOcean isn't interested in making money as a business, they're happy to lose money as long as they get more in the front door from investors. Meanwhile other low end providers look more like lifestyle businesses and thus growing slowly and organically.
Except for that part where as they grow they develop economies of scale that allow them to undercut the "lots of providers". There's more than one way to play the game.
I just don't feel there is anything special about DigitalOcean, nothing that indicates to me that the founders really care about the technical nuances and details of server hosting.
Instead, my impression is that the company will continue to be pimped and supported until a profitable exit is reached. It seems to be all about money. It's hosting today, but tomorrow it could be nursing homes.
It seems to me that economy of scale does not translate so well to hosting companies. Sure you might be able to get slightly better pricing on servers and connectivity, but complexity has a cost, and it goes up as you get bigger.
And would you trust your hosting with Millions of VC backed money or a small VPS company? ( No offence, it is all part of perception and marketing )
I think for less then $10 hosting only, Ramnode and many others LEH wins. But if you want sometime in production that needs more power in the future DO have you backed.
I don't see any reason to "trust" DigitalOcean more or less than a company like RamNode. VC backed money is not a guarantee against failure or problems with your hosting account.
The potential of instantly scaling up to 96GB of memory and an API are an advantage, but it's not exactly a killer feature. I would imagine it's a small minority of DO customers that ever use those features.
2. Exceptional user experience
Quoted metrics for support for ARP and MAC addresses on devices didn't match what we were seeing in production but we were able to get that resolved, so you shouldn't see those issues recurring.
1. Private Back End Network.
2. Auto Provision Droplet/Instance on different hardware by default
4. DC in Asia
5. Something Similar to Amazon Elastic IP.
Things i would like
More Powerful CPU
Higher Quality Network ( Although it has gotten a lot better in recent months )
combined costs currently at $140/month
each app has 1 free web dyno, 1 worker, starter postgres DB, ssl, plus a few extras
Looking at the new offerings like digital ocean, i'm really tempted to switch over, but a voice in my head keeps telling me it makes no sense as I don't really have strong linux-sysadmin type skills. Even getting rails to work on new macs takes me 1-2 painful days.
Anyone have an eta of what it might take someone with limited sysadmin skills to cut over to something like digital ocean from heroku?
Feel free to read the comments for yourselves: http://reviewsignal.com/webhosting/company/101/digitalocean
They are obviously doing something right and a huge congratulations for their fund raise.
Privacy violations would make liable under the local (a western european country) Data Protection act
I hope they can add more features while staying price-competitive.
BTW, I'm shocked that this round was SEED. What happens when they get Series A. Will they continue to change game rules? I think yes.