Microsoft Has Created Its Own FreeBSD (microsoft.com)
Simon Sharwood, writing for The Register: Microsoft has published its own distribution of FreeBSD 10.3 in order to make the OS available and supported in Azure. Jason Anderson, principal PM manager at Microsoft's Open Source Technology Center says Redmond "took on the work of building, testing, releasing and maintaining the image" so it could "ensure our customers have an enterprise SLA for their FreeBSD VMs running in Azure". Microsoft did so "to remove that burden" from the FreeBSD Foundation, which relies on community contributions. Redmond is not keeping its work on FreeBSD to itself: Anderson says "the majority of the investments we make at the kernel level to enable network and storage performance were up-streamed into the FreeBSD 10.3 release, so anyone who downloads a FreeBSD 10.3 image from the FreeBSD Foundation will get those investments from Microsoft built in to the OS."
The interesting thing is that you would never see this happen under previous leadership. Forget the Windows 10 mess, even forget Microsoft selling one-off software at all. They are absolutely committed to using Azure to become the next IBM. The reason why IBM is still alive is because they draw massive monthly revenue from the mainframe business. You don't just buy a mainframe and a z/OS license as a one-time thing. You buy the hardware, the licenses, plus a huge monthly maintenance charge, _plus_ a pay-by-the-MIPS charge to use the hardware. IBM maintains the system for you, sends minions to replace parts, gives you access to upgrades, etc. for this fee. In an environment like this, it makes perfect sense to allow customers to run whatever they want as long as they run it on Azure. Microsoft will be the toll collector for anything their customers choose to migrate there. I'm working on a big Azure migration/rebuild project, and it's so obvious that Microsoft is done pushing their own software...as long as you rent their infrastructure.
Don't know why you're getting the downvotes.
I agree with you on systemd. Linux did need full process management. I do like how systemd standardized the init system and it'd be nice if there could be a simple drop-in replacement for it. Uselessd development has stopped and no one has the time to contribute anymore. The big projects are all funded by the big for-profit companies.
I use Gentoo and still love it. I don't use systemd, but it is an optional choice (like it should be). I haven't touched FreeBSD in years, but I can understand people moving that route. I might load it up at some point. I just don't have the time to invest these days and Gentoo still works great for me at work and at home.
You know how I know that you have no real understanding of IT?
I'm no Microsoft fan, but, they did submit their changes back upstream to FBSD. The BSD license doesn't require them to do that, but they did.
file:
Don't spell it "systemd".
Spell it "SystemD"
That way it looks like an ASCII penis.
Microsoft wants Linux to work better on Azure
I'm not sure I'm following. Microsoft made some changes to FBSD, so it will work better on Azure and Hyper-V. They submitted those changes back to the main FBSD project. What does any of that have to do with Linux?
file:
Hi folks,
Disclaimer: I'm a FreeBSD committer.
MS has been committing various Hyper-V drivers for months. Just like VMWare does for its hypervisor.
This is less
and more
You know, like every other cloud vendor's VM images. Nothing to see here, move along.
So, stop Hyper-Ventilating! ;-)
Having watched Windows grow from MSDOS, to Windows 3.1, to NT and beyond, and having observed the architectural stability through those stages (e.g. the registry), I have become convinced that the only way Windows will become truly stable and easy to maintain will be for it to adopt a UNIX architecture. This is not an absurd suggestion. Apple did it. It adopted a UNIX kernel, and managed to support legacy programs using virtualization. The process was in fact relatively benign from a user point of view. Old software appeared to continue working as it used to. When one opened a program written for legacy MacOS, a virtualized environment was created, and the program worked in the same way as it did in the older OS, even though it was actually running in OSX.
Windows could do this with relative ease. Create a brand new OS on a UNIX foundation. Create a virtualized environment to run legacy software. The god damned registry and all the other architectural mistakes can live in that space as long as MS wants to preserve legacy support. In the mean time, MS can move on from the detritus that has built up in Windows over the years. It can have a fresh start. The new windows can have things that other UNIX operating systems have enjoyed for years, like for instance proper hardware abstraction. Imagine using the same OS foundations on phones and laptops, like Apple has had for years. Imagine supporting different processor architectures with the same basic OS, like UNIX systems like OSX have had for years.
I left Windows years ago, partly for the reason that the OS was so badly engineered. It shocks me that Windows still runs the software engineering abomination that is the Registry. I currently use a combination of OSX and various UNIX systems. I will NEVER return to windows unless MS upgrades its OS to a more stable foundation.
This and no other is the root from which a tyrant springs; when first he appears as a protector - Plato (423 to 327 BC)
SystemD is happy-path programming at its finest. Just don't expect to fix your system when something goes wrong. SystemD's failure paths violate Principle of least astonishment. I work with these type of people all the time. They test the heck out of the happy-path and everything works great, but if they can't test it, they can't imagine it, so they have all kinds of corner cases, and the strange assignment of responsibilities makes the system difficult to reason about.
Programming skill is distributed on a power curve. SystemD's design is below average, but well above median, so I guess it's an overall win.