Who's the main target for Azure? Enterprise companies who trust Microsoft implicitly. When an exec comes to the head of IT and says "we need to be on the cloud! I read about it!", Azure eases the transition by being able to go to vendor you've already been using for a dozen years.
RedHat's core audience is enterprise as well. RHEL is the de facto standard for that level of infrastructure due to the support you can get versus any of the distributions that are equally as good, but minus the support contracts.
So, they're helping each other out and that's good in my humble opinion.
Microsoft's new direction under their new CEO is one surprise after another. I've only tinkered with Azure so far but makes me want to pay MS more attention than I would have a couple years ago.
I don't know about that, it all seems fairly logical. Don't forget that Microsoft is primarily a "business business" rather than a software business (ie. optimizing for business longevity above all else).
Microsoft appears to have practically perfected the game of "maintaining vendor lock-in in an otherwise open ecosystem". That game requires "giving in" at times, when it is no longer viable (in the long term) to compete with other options.
A good example of this is the open-sourcing of .NET. I don't believe for a second that this was a change of heart of a developer or even a team - it is far more likely that Microsoft is realizing the increasing shift away from Windows and .NET, and towards more 'open' platforms (whether Python, Node.js, Ruby, Golang, or whatever else).
It is in their best interest to make .NET open-source, as it allows them to maintain their foothold in the application development community - it still has direct integration with the rest of their products (and thus incentivizes picking MS as a vendor), but can now compete on openness.
You can see something similar for Windows 10, and it being given away for free. Both OS X and Linux are increasingly eating away at Windows' marketshare. By offering it for 'free' to existing Windows users (ie. nearly everybody), they can attempt to win users back, as it is now offered at the same price (in the eyes of the consumer).
Throughout the existence of Microsoft, they have consistently pushed the boundary of vendor lock-in and marketshare, trying to keep it as closed as possible but as open as necessary. The more recent decisions from Microsoft are not surprising to me at all - they are simply the result of a rapidly changing computing landscape. Microsoft hasn't changed, their environment has.
For people dealing with Microsoft for many years... this is the surprising part.
Microsoft open sources their languages and then buys out the company that does the best job in making it crossplatform with as many platforms as possible.
What Microsoft can't do they usually buy out a company that can do what they can't. Getting Visual Studio for other platforms is near impossible for them, so they open source the CLR and compilers and see who can do a better job at porting them to other platforms.
Microsoft makes a lot of money selling Workstation and Server version of Windows to businesses as well as Visual Studio licenses and BackOffice (SQL Server, Exchange Server, ISA Server, Sharepoint, etc) for Windows Servers to make custom apps on. Not to mention Office licenses and other software for businesses.
New CEO got Microsoft into the Cloud and Cloud services, so it only makes sense to make a deal with Red Hat for an Azure Server deal to sell to more businesses that want a GNU/Linux solution with paid support.
Since Ubuntu ruling the cloud instances this also makes sense for Red Hat.
It would be cool to get numbers across all providers, in terms of deployed instances of this or that OS. There's a good story to write, there.
That's were Canonical began to gain traction regarding installs, yet they still can't find a way to get users into support contracts en masse like Red Hat has been able to.
Citation needed. If Amazon Linux is still the default, you're wrong. It is based on CentOS, uses RPMs and yum.
(And I edited my comment since the wording was unclear... sorry if that caused confusion)
Ubuntu has gotten some solid traction in the "web" end of things, while RH is more "infrastructure".
And different Linux flavors were already on Azure as "applications" prior to this deal.
But now RedHat is getting the official stamp.
If Azure the OS only had Windows the App, that would be a great failing. MS is smart and playing Azure out to not just promote other MS products. Expanding it as an ecosystem for various OSes is a Good Thing.
I say this as someone who works for an ISV that is single-handedly responsible for the only Linux servers most of our customers have.
This was the biz model for LinuxCare.... but they failed.
That's the point of a partnership.
They are using Azure as a loss leader to get you hooked, then the price goes up. This new SaaS world is kind of like buying cars.
That said, nothing big, and planning to migrate things to linux solutions but time and motivation don't always align for personal stuff at the end of the work day.
They seem to be one of the better options if you need hosted windows anything as part of your cloud strategy. And having played around with Azure's other services , it seems to work pretty well all around. I did find the default interfaces for Azure's storage a bit difficult to work with, so I created a wrapper with imho a nicer interface.
1) The reseller doesn't know how it runs over Azure (I had this with vendors sticking to in-house Sharepoint vs Sharepoint online mostly from a "we don't know how" angle we found out)
2) Does the reseller also make money from in-house vs on Azure? Like, their hardware? Anything like that? Just curious.
I mean, he could also honestly think that Azure isn't there yet either. I could just be curmudgeonly and doubting the reseller.
I've seen folks spend 3-4x more money (almost all of the excess to the SAN vendor) to get a VMware setup (because snapshots and VMotions) with the same reliable I/O a commodity Dell server with a 4-disks SAS Raid-10 and a decent RAID controller can deliver, not even talking SSDs here.
We always said to use "the right tool for the right job", isn't it? There are a lot of apps for which virtualized infrastructure is just not a good fit from a technical standpoint. When these end up in the cloud because suits decided OpEx is sweeter than CapEx, they suffer horribly; the customer ends up paying more (because prolonged I/O spikes in the cloud can be hugely expensive), the application runs slower than it did 5 years ago on commodity hardware, and nobody is happy.
Still, suits decided cloud it must be, so cloud it will be. Sometimes the IT sector is depressing.