So last year I livetwote my reading of the @rightscale / @flexera state of the cloud report. I do it again this year in the hopes that it's better than the dismal showing at the taunting of @LawrenceHecht. Blame him.
We begin with the executive summary on page 9.
"93 percent of enterprises have a multi-cloud strategy; 87 percent have a hybrid cloud strategy." That seems awfully high, I wonder what @flexera does?
It might be multicloud, but to be honest their site is incredibly vague.
"93 percent of enterprises have a multi-cloud strategy; 87 percent have a hybrid cloud strategy." That seems awfully high, I wonder what @flexera does?
It might be multicloud, but to be honest their site is incredibly vague.
"For cloud beginners, lack of resources/expertise is the top challenge; for advanced cloud users, managing cloud spend is the top challenge"
Yes, you've captured the real cloud journey quite succinctly.
Yes, you've captured the real cloud journey quite succinctly.
"VMware vSphere continues to lead in private cloud adoption, with Azure Stack and AWS Outpost showing the most vigorous growth"
...this time last year AWS Outposts weren't available yet. I bet its growth from "zero" was freaking tremendous, professor!
...this time last year AWS Outposts weren't available yet. I bet its growth from "zero" was freaking tremendous, professor!
Okay, I'm no geographer or anything but where pray tell is the "rest of world" referenced here? 
The problem, as always, with this multi-cloud narrative is definition in terms.
Basically nobody is all-in on AWS to the point where they'll eschew Zoom, G-Suite, GitHub, and Dropbox. And if they are, they're pretty goddamned miserable.
Basically nobody is all-in on AWS to the point where they'll eschew Zoom, G-Suite, GitHub, and Dropbox. And if they are, they're pretty goddamned miserable.
Okay, THIS is more like it. Yes, different workloads in different clouds. That's what a human being with half a functioning brain does in the common case. There are exceptions FOR SPECIFIC WORKLOADS, but that's where it ends.
"Most Multi-Cloud Management Tools Are Complete Dogshit" would have been the more accurate graph title.
And note the editorializing of "unfortunately." I guess @flexera sells a multi-cloud management tool!
You'd think I'd have found that out from their actual website.
And note the editorializing of "unfortunately." I guess @flexera sells a multi-cloud management tool!
You'd think I'd have found that out from their actual website.
This graph tells us more about the survey respondents than it does the industry. If I poll my customer base the numbers look... different. :)
Here Flexera helpfully teaches us that small companies spend less on cloud than big companies.
"SMBs are adopting public cloud faster than enterprises" they further state.
SMBs do basically everything faster than enterprises, up to and including "evacuating the building during a fire drill."
SMBs do basically everything faster than enterprises, up to and including "evacuating the building during a fire drill."
"Dependency mapping is top cloud migration challenge" comes as a surprise to nobody who's ever tried to map out even a small microservices architecture.
And here I pause to go check on a toddler who has awakened.
I return!
Strangely "colorblindness" doesn't appear on this list of challenges, but it definitely should. Shades of blue...
Strangely "colorblindness" doesn't appear on this list of challenges, but it definitely should. Shades of blue...
"Enterprises are gaining experience with cloud, which has led to slight decreases in perceived cloud challenges."
Yup, ask any enterprise. "CLOUD IS SO EASY!" they all cry joyfully.
Then we ride unicorns and eat ice cream.
Yup, ask any enterprise. "CLOUD IS SO EASY!" they all cry joyfully.
Then we ride unicorns and eat ice cream.
"Are you predicting your org's cloud spend will grow in the next twelve months?"
Let me spoil it for you: it will, barring a meteor wiping out northern Virginia.
Let me spoil it for you: it will, barring a meteor wiping out northern Virginia.
And now we get into the meat of it!
I can't talk about a lot of this, so ask me no questions and I'll tell you no lies, but this chart blows.
"Ad hoc discounts" my ass.
Only 53% of respondants are using RIs?!
It's anecdote time...
I can't talk about a lot of this, so ask me no questions and I'll tell you no lies, but this chart blows.
"Ad hoc discounts" my ass.
Only 53% of respondants are using RIs?!
It's anecdote time...
By the time you're spending over a couple of million bucks a year, you're not paying retail on virtually any cloud provider.
Some people are going to fight me on this. I'm used to it. I've been told emphatically by engineers that they don't have any discount deals in place--at companies where I helped negotiate their discount deals. Not everyone in every shop knows what deals they have.
I digress.
I digress.
This chart is, to no great surprise, also complete garbage.
I think I use most of these *IN MY OWN ACCOUNT* for different projects. Instead this report paints huge complex orgs with an overbroad brush.
I think I use most of these *IN MY OWN ACCOUNT* for different projects. Instead this report paints huge complex orgs with an overbroad brush.
"Are you using Kubernetes?"
"No, we're using EKS."
"No, we're using EKS."
AWS Outposts have only been out six months, and already half of the respondents have one? Sure, @flexera. Whatever you say.
Not pictured: respondents who say "what the hell are you banging on about? 'Private Cloud' is a contradiction in terms, we run goddamned data centers."
We close with a rousing call to action that still doesn't tell us a single freaking thing about what @flexera does past putting out hilariously inaccurate reports.
In conclusion, I trust @flexera's report to give me accurate insight into the industry about as far as I can comfortably spit a rat.
Do better or stop doing these, please.
Do better or stop doing these, please.
