IT Procurement 101: AWS or Bust
As some of you may know, DropBox left AWS and developed its own cloud technology, including both hardware and software. They realized that a deep commitment to AWS did not make business sense in the long-run. They also learned that leaving AWS would be a technical and logistical nightmare.
Cade Metz has an article in Wired that tells the DropBox-AWS story. He notes the logistical problem:
A project like this, needless to say, is a technical challenge. But it’s also a logistical challenge. Moving that much data across the Internet is one thing. Moving that many machines into data centers is another. And they had to do both, as Dropbox continued to serve hundreds of millions of people. “It’s like a moving car,” says Dan Williams, a former Facebook network engineer who oversaw much of the physical expansion, “and you want to be able to change a tire while still driving.” In other words, while making all these changes, Dropbox couldn’t very well shut itself down.
For the full article see http://www.wired.com/2016/03/epic-story-dropboxs-exodus-amazon-cloud-empire/
=======================================================
Comment: I spend countless hours each week researching and analyzing vendor success and failure to help ensure my customers are making the right business decisions. This article was a breath of fresh air and I hope others can learn from forward thinking companies like DropBox. AWS is not a quick fix and just because your competitor is using them doesn’t mean you should follow along blindly…frankly, moving in a different direction may be an advantage.
I recently spoke to the head of procurement at a fortune 100 company notorious for sending out an RFP for everything. “Cloud?” he said, “Our CTO has made an executive decision to skip an RFP and just put everything into AWS.” I nearly fainted. The decision was made for two related reasons: 1- the sheer complexity of a large-scale cloud or hybrid infrastructure could mean a project that takes years, and 2- if they didn’t get onto the cloud quickly, they would somehow be losing out to competitors.
Think back 5 or 6 years to the early adopters, the innovators and forward-thinking companies that chose to put their environment into AWS in the early days. These are the now the companies trying to find a way to migrate large portions of their environment out of AWS due primarily to the horrible economics. While migrating out of AWS would not be easy or cheap feat, once accomplished it could reduce their monthly spend by 20-30%... serious bottom-line revenue for most companies.
Late adopters in spaces like healthcare, publishing, education, etc. who are rushing to get into AWS need to put on the brakes and learn from the early adopters. AWS is good for some things, but not everything. AWS is not a quick fix and, once committed, it can be hard to change. Remember, the future is in hybridization and running applications where they are best suited, technically and economically.
Also as SDDC, XaaS, HyperConverged, Automation and Orchrestration technologies improve over time, the TCO & ROI equation for private / hybrid clould will get better, futher reducing the cost gap compared to public cloud for increasing number of workloads within Enterprise IT. Every business needs the agility of being "Cloudlike" ....which form of cloud they choose depends on when they decide to take the plunge
Agree with your comments Greg. We talk to a lot of IT teams that are under pressure to a cloud strategy, but fortunately logic seems to prevailing in most IT shops. Some workloads are a good fit and some are not, with the decision largely based on cost, performance, and control. For example tier 1 enterprise apps that keep the business running, are applications where risk mitigation and management of the end-user SLA is far more important than saving a few bucks. [thinking of some Yoda quote about "...taking the easy path..."] But I'm also a believer in "hybrid"...*not* that a single application will span public and on-premise resources (a very rare business requirement), but across the hundreds of applications running in an enterprise, we will find a logical disposition across public, private, and pseudo-private (MSP) cloud execution environments. To get there, the industry has a long ways to go to reduce friction. Friction of migrating, friction of deploying and managing on-premise technologies, friction of complex learning curves, friction of incompatible APIs. Lots of room for innovation out there.
Well said "Remember, the future is in hybridization and running applications where they are best suited, technically and economically." Sadly, it seems like a full IT exit was needed to reset economics (?!) but once reset, there are new options to consider.