Hi, I’m Steve Saunders. Two decades ago, I founded a site called Light Reading to cover the boom in optical networking technologies. Now I’m doing the same thing to focus on an even larger, even more important market.
The new site is called Silverlinings. It goes live on Jan. 5, 2023, and it focuses on cloud infrastructure.
Note that second word in there: infrastructure. It’s important.
The last decade has seen a horrible trend for cloud-washing across industries, where companies just add “cloud” to whatever the heck it is they’re selling, because cloud is perceived as a hot, must-have marketing buzzword.
Silverlinings doesn’t cover cloud in a generalized, hand-waving, skip-over-the-difficult-bits kinda way.
It covers cloud infrastructure — the technologies that enable all those wonderful and vital cloud services and applications to work. And it covers the companies that develop and sell that technology. And it covers the pioneers that are using it.
And it covers the long list (and I do mean long) of challenges they face, from the lack of standards for cloud, to avoiding hyperscaler lock-in, to the demand for certification and training, to meeting reliability, compliance and regulatory commandments using open-source code to... oh well, just a whole bunch of other trying stuff.
Cloud network architects: looky here
Silverlinings' coverage is written with a very specific audience in mind: the individuals who’ve been tasked with deploying cloud infrastructure within telcos and service providers, enterprises and key industry verticals (finance, healthcare, automotive and so on).
We’re calling those people “cloud network architects,” though their actual job titles vary wildly.
I’ve spent the last three months talking to makers and users of cloud infrastructure technology, and my biggest takeaway is that there is a massive disconnect between those cloud network architects and the marketing around cloud infrastructure. Today, cloud messaging is focused on how exciting and cool cloud is and how easy it makes things.
But the No. 1 emotion from the people deploying this technology is ... anxiety.
Oops.
Why is this? A big reason is that cloud networks are really, really complicated. And that complexity is made significantly worse by the lack of any commonly accepted definitions around cloud infrastructure (something that goes hand in hand with the lack of cloud infrastructure standards).
Most people in our industry can draw a cloud network topology on a white board. But ask them to label the products that support that topology, and the answers are all over the place.
The problem is that there isn't a universally accepted reference model defining the products that support the cloud. (I know, dumb.)
Heck, right now we can’t even agree on what the different elements of a cloud infrastructure should be called!
In the absence of an agreed-upon nomenclature, vendors are just calling their products whatever they want. There are an awful lot of “solutions” out there. This is very unhelpful. (Note to marketing types: I’m writing this on a solution, while sitting on a solution, after I took the solution to work this morning...see? Annoying. Stop it.)
And “gateways” are another popular but crap term. (Turns out a lot of those datacenter gateways you’ve been hearing about are routers with a big dollop of cloud code sitting on top. Maybe just say that, people?)
I get why this is happening.
In the old days, network infrastructure came in a box. Literally. We gave the different boxes names, and we knew where they went in the network. You could actually, like, see the technology. (“Hey, there it is, over there!”)
Hell, you could give it a kick when it wasn’t working or pat it on the top of its enclosure if it had done its five-nines duty. (“Who’s a good CRS-1? Yes, you are!")
But with cloud infrastructure, the whole game has moved into software, and it just refuses to stay put (that whole “virtual” malarkey). It might be on a blade or a standalone server, on top of a rack or in the spine, or a leaf switch. White box? OK. No box? Fine! Distributed? If you like. Centralized? Sure. Hardware? Yes! But also ... no. Maybe?
Geo-locating the code is just the start of the challenge. Cloud infrastructure software products perform an amazing mixture of tasks.
You be the customer; I’ll be the vendor.
Let’s role-play the product evaluation process. You be the customer; I’ll be the vendor. (Note: You have to actually say this out loud or this doesn’t work.)
You: “Hullo, I’m a cloud network architect, and I’d like to buy a product to manage my cloud, please.”
Me: “I’m sorry sir/madam, could you be a bit more specific? Would you like some cloud networking management or cloud data management? We have some cloud edge solutions on special discount this week, and I think there might be some container and Kubernetes management in the back. Or is it cost management you desire? Or hybrid cloud management? Security management? Cloud-based database management? Well? Eh? Hmmm? Speak up!”
And that’s just cloud management. Don’t get me started on cloud integration.
And, of course, lots of these products do more than one fundamental task — management and cloud networking all on a sesame seed bun.
All of this irks me and offends my orderly nature. As a communications analyst, I demand that my communications products fit in a well-arranged table with plain English headings for their key features so I can compare them. But today, there is no table or headings for cloud network infrastructure.
So Silverlinings is going to make one. A really BIG one. And we’re going to need your help.
We’re developing a cloud infrastructure taxonomy (working title: Silverlinings Playbook).
We’ll be launching it in Q1 2023.
In addition to being a definition tool, our taxonomy will also be a directory.
So, if you look up a term, you will be able to see every company that provides a solution in the space and what they sell. And while we're at it, we decided we may as well add people to our taxonomy/directory. And a dictionary of industry terms.
So, n-dimensional.
In January, we’ll be issuing a call to action to the industry to submit their cloud infrastructure products and services for inclusion. But before that, we have to get our terms and definitions straight. That’s where you can help.
I’m looking for some cloud industry-types to help define the taxonomy dataset. Keep in mind that this is not a cloud taxonomy — it's a cloud infrastructure taxonomy.
Silverlinings isn’t written for application developers. And we don't write for consumer users of services that are delivered over a cloud. We write for cloud network architects. So please only reach out if you want to contribute ideas around categories or products or features that are relevant to our audience.
This is important work, and we need your help. Shoot me an email at [email protected] if you want to be involved.
See you in January!
P.S. Sign up to receive the Silverlinings newsletter here. It will ship out every Thursday.