The EVIL Cloud strikes again
As I’ve been saying for 2-3 years now, Cloud Technology is still WAY too early for any company to be using as their “end-all” strategy. And for sure, Public Clouds may be good for Mom & Pop organizations or for large inefficient ones like the US Federal Government (to be sure!) but for any major corporation or even a new fast Startup to throw their critical infrastructure onto Amazon, Rackspace, Google or any other Public Cloud and walk away expecting nothing but rainbows and pretty pictures, is foolhardy at best and downright idiotic and a fire able offense (in my opinion) at worst. I’m sure many board of Director’s are screaming bloody murder the past week and would not be surprised to see some heads rolling.
The problem is that ANYONE who has been in this business 10 years or longer KNOWS (or should) exactly what will happen. Anyone who has had a hosted server, Hosted DNS, or Hosted Website for ANY length of time KNOWS they eventually go down, they all do. So why ANYONE with any experience would place their core business and their whole company in that situation is beyond my comprehension.
The only thing I can think of, as I’ve also stated previously and have seen first hand, is that a lot of these new startups and I consider Reddit, Foursquare, Pinterest, etc… ALL as young and inexperienced startups with Young Executive staff (I’m talking ALL in their twenties) and a refusal to hire older, wiser heads to help them with their business and infrastructure strategies. You may think I’m full of hot air here, but believe me I’ve SEEN IT MANY MANY TIMES over the past few years folks. I’ve spoken till blue in the face in many cases and in one my 25 years of experience was over-ruled by a 21 year old software developer who thought he was an infrastructure specialist and demanded the company use Amazon Cloud Services. Even after a two week outage at Amazon almost lost them their company, the CEO who was also around 21 years old was apparently buddies with this software developer, and even though the one older person they had working there (the VP of Software Development who was in his late 40’s and was the one who recommended they call in me) also advised them to migrate to their own Private Cloud, they decided to ignore our combined experience of close to 50 years and doubled down with Amazon and Rackspace.
I cannot tell you how many new startups I’ve interviewed with who have fallen for the “marketing hype” and believe that Amazon is the worlds gift to Cloud and some (I’m not kidding here) even had no idea there was anything other than Amazon! As I’ve said many people equate Cloud=Amazon and so in my Not so Humble Opinion, that is the major reason many of these companies are in trouble and some are bailing ship and finally deciding to build their own private cloud infrastructures and I would bet you anything they finally brought in someone like myself our “Cloud Consulting International” who finally convinced them, as every technology analyst has written reams about (ie, Gartner, IDC Network World, and others) how the Public cloud IS NOT SAFE and WONT BE FOR 5-10 YEARS YET for any sort of “critical” infrastructure or “sensitive” data and that the majority of businesses other than MoM”n”PoP shops need to build their own private cloud infrastructures FIRST and then SLOWLY, in the future, perhaps they will be able to migrate some of the infrastructure to the Public cloud.
Spiraling out of Control
What has happened with Amazon, EACH TIME that they have experienced an outage, something small has spiraled out of control into a Major Catastrophe. That is what typically happens when your relying on minimum waged (or not much higher waged) employees to manage a complex infrastructure. I’ve seen it during the past 18 years in almost every hosting facility/colocation facility I’ve worked in or visited. The fact is you have a massive amount of square footage and many hundreds of thousands of pieces of equipment if not millions in these large public hosting companies, OF COURSE they cannot afford to hire skilled workers like myself or others to manage all the little details, they’d go broke, so they hire college kids with no experience. You would be surprised.
So add it all up, massive infrastructure of millions of components, a NEW technology “cloud”, college aged kids with no or minimal experience running the place and in many instances in upper management even, A Massive Marketing Hype and Advertising Campaign telling everyone that their service is 100% Safe, Secure and Reliable (They still claim a 99.99% UPTIME! Which is a physical and scientific impossibility with the number and length of outages they’ve had the past 3 years, even a single hour outage can affect uptime percentage for months) and the Best Thing Since Sliced Bread and you have the makings for continued disaster after disaster.
My Best advice?
Do yourself a major Favor, Call “Cloud Computing International” or other respected cloud consulting company or hire yourself someone with at least 15 years of data-center and/or Cloud Infrastructure Experience BEFORE making any major commitments to your core infrastructure and the cloud.
Believe me, If all I was after was money, I’d be whitewashing this over like many others are doing so as not to kill the goose with the golden egg!
- Startup claims it saw early signs of Amazon’s cloud outage (techworld.com.au)
- After Amazon Outage, Rivals Seek to Capitalize (datacenterknowledge.com)
- Amazon Web Services outage once again shows reality behind “the cloud” (arstechnica.com)
- Amazon Cloud Outage KOs Reddit, Foursquare & Others (datacenterknowledge.com)
- Amazon cloud outage brings Reddit, Pinterest, Netflix down (androidcommunity.com)
- Amazon outage takes down Netflix, other sites (nbcnews.com)
- Amazon Web Services Failed, but AirBNB Didn’t Have To (cloud.dzone.com)
- Amazon Cloud Outage Affects Major Web Properties (coated.com)
- Reddit, Flipboard, ‘Minecraft,’ and others taken out by another Amazon Web Services outage (theverge.com)
- Amazon outage takes down Reddit, others (cnn.com)