Microsoft has developed its own Linux. Repeat. Microsoft has developed its own Linux Sitting down? Nothing in your mouth? Microsoft has developed its own Linux distribution. And Azure runs it to do networking. Redmond's revealed that it's built something called Azure Cloud Switch (ACS), describing it as “a cross-platform modular operating system for data center networking built on Linux” and “our foray into …
U-ho.
Their mobile efforts also failed, so their only way forward was infiltration. Is this the way we'll see Linux on the desktop then?
Maybe we could ask cortana.
I can't see Microsoft's version of Linux being any better than their own desktop software.
What I can see are patient and copyright lawsuits by MS against the established Linux world.
The DoJ needs to revisit the breakup of Microsoft into Office software and operating system software.
On the contrary, as this software is based on Linux, which is GPL'd - remember why MacroStuffed hires hundreds of corporate lawyers? - I'm eagerly awaiting my copy of all of the source code for this software, because the GPL requires that, and Microsoft hates software piracy and people who don't comply with every letter of their software licence, doesn't it? :)
> I'm eagerly awaiting my copy of all of the source code for this software, because the GPL requires that,
The GPL only requires that the source code be available to those who receive the software. As the software is only distributed internally then there is no requirement for source code to be made available outside.
Presumably they only did this because it was cheaper and easier than recompiling Windows for these ASICs. Surprised the performance is good enough though. As anyone who has tried really high end connectivity like Mallanox or 40GB Ethernet, the Windows network stack significantly outperforms off the shelf Linux - which doesn't support features like TOE without custom kernel hacks.
"Surprised the performance is good enough though."
Why? For starters the OS doesn't move frames in a switch, it's just there for abstracting management of the actual ASIC(s).
"As anyone who has tried really high end connectivity like Mallanox or 40GB Ethernet,"
That would be me. And in reality it depends on many more factors than just the OS, such as NIC chipset and driver quality. For many day-to-day tasks I've seen little discernible different between Windows and Linux at 10G or 40G. I've found Linux (and *BSD) often pull ahead slightly once you increase the frame size (who doesn't these days?) but in reality it's hardly enough to bother writing home about.
"I've found Linux (and *BSD) often pull ahead slightly once you increase the frame size (who doesn't these days?) but in reality it's hardly enough to bother writing home about."
Not my experience - high bandwidth / low latency networking is almost always faster on recent Windows versions than Linux (can't comment on BSD) - examples I have seen tested being as an NFS server - and as a SMB server (using SMB Direct) it's way faster than any Linux option I have ever seen. Does Linux have an SMB Direct driver yet? The Samba team were "working on it" 2 years + ago...
"Why? For starters the OS doesn't move frames in a switch, it's just there for abstracting management of the actual ASIC(s)."
That makes more sense, thought this was the OS routing the packets. Which Windows tends to do very well these days due to the fully modular network stack, and better native hardware integration - for instance TOE and RDMA support is a native part of the OS, not a bolt on.
Or even "Microsoft Zune", another typical microsoft catch-up "me too" failed experiment, fit only for posters on tech forums to mock by constructing fake names from.
Ethernet switches have been using Linux-based firmware on the inside for ages - especially the lesser known brands / switch vendors. Cisco traditionally had their own in-house IOS, but I seem to recall that some more modern IOS strains on some HW platforms are actually linux-based too... Other popular operating systems to use for firmware are the various BSD flavours and various RtOS'es (QNX, VxWorks and the like). The CPU cores used in switching hardware (= what actually runs the firmware code) are typically PowerPC, ARM, or MIPS - Linux supports all of them. If the Ethernet switch chipset makers provide some reference firmware platform, it will most likely be Linux. So if someone like Microsoft possibly decides to develop their own firmware for some 3rd-party OEM switch hardware, Linux is a very logical choice. That's where they're likely to get the best technical support, needed to bootstrap Linux on the management CPU core, and in terms of drivers and API's for the specific hardware (L2 switch matrices, L3+ accelerators, DMA engines, individual mac/phy blocks, various misc IO such as I2C/SPI/GPIO). But I still consider it a little unlikely that they're going all the way from bare metal (Linux from scratch). I would find it more natural if they took whatever reference firmware (Linux) the chipset maker has provided, and port the Microsoft's own user-space tools / API's to it, while possibly bugfixing and modifying the reference firmware a bit in the process.
So it is not cats sleeping with dogs, Lucipher driving a snowplough or the four horseman riding out.
So while an improvement (in Balmer days they would have twisted Broadcom's arm to stick Windows on it or gone for vXworks), it is out of necessity. It is not a result of finally having a more adequate perception of reality and the actual merits of Linux vs Windows for real work.
"MS using an appropriate tool for a particular task and adapting it."
But that *is* an interesting story.
Imagine if the idea of "using an appropriate tool for a particular task " caught on elsewhere.
Certified Windows dependent IT departments around the world would be up the creek without a propulsion object.
IT projects might succeed (according to organisation-defined not IT-defined criteria), and sometimes success might even cost less than today's failures cost organisations.
Bring back Windows CE, before this embedded Linux insanity catches on in Redmond.
The end is nigh.
"IT projects might succeed (according to organisation-defined not IT-defined criteria), and sometimes success might even cost less than today's failures cost organisations."
IT Projects succeeding or failing usually has very little (in fact nothing) to do with the underlying Operating System. Also, IT defined criteria *is* the organization defined criteria. Not following? Ok.
Junior person in hospital wants to transmit pictures of abused kid who's come into the hospital to social services so they can investigate if the parent is abusing the kid. Junior person specs the requirement as:-
1) "I want a computer with an internet connection so I can email a picture of an abused kid to Social Services".
Simple, right?
IT requirement for the same job probably reads something along the lines of requirement "method of making abuse report and securely transferring images to Social Services".
1) We must be able to prove that we have sent it.
2) We must be able to prove that they have received it
3) We must be able to prove that we have made the report in a format that can be input to their internal systems.
4) We must be able to prove that they have actioned the report to fulfil our duty of care to the patient.
5) We must be able to prove that the images have been sent securely.
6) We must be able to provide an auditable trail of who has seen, received, discussed (etc) the child and said pictures in case Social Services takes the case to court and demands that our staff give evidence.
In the hypothetical situation above, Do you really think that IT put all of the other requirements in for their own amusement, or do you think the IT people were tasked by the hospital to ensure that their IT systems follow all of the local requirements like the hospitals duty of care, privacy of the people involved, governing body requirements, laws, etc?
"Do you really think that IT put all of the other requirements in for their own amusement,"
What you wrote makes a great deal of sense.
Unfortunately in my twenty odd years experience of IT departments (as a non-IT employee with a clue about IT) less than 3% of IT departments and IT people I have experienced would have a clue about the perfectly reasonable stuff you wrote (some of which arguably are user requirements not IT requirements anyway).
IT requirements:
1) Does it run on Windows?
2) Do Microsoft sell it?
3) You're in.
@A/C all I feel is pity for you, most of the (other) people I work with would operate at that level, its what we get paid for.
Personally it sounds to me like you have a very low opinion of IT professionals, or you've always worked alongside the low hanging fruit.
(some of which arguably are user requirements not IT requirements anyway).
With regards to user requriements: i'd agree that what I tacked on there are in fact user requirements, but that's because the users rarely know what they are actually asking for and never check that their proposed solution is actually acceptable to the business.
Does it fit in with existing processes for their department, other departments and external entities? They probably don't know their job is a carefully designed little cog in a bigger machine, let alone consider how it fits in, or that if they provide two extra bits of information initially it saves somebody else 5 telephone calls trying to get hold of them.
Does the solution meet local laws and requirements from the governing body? Users simply don't consider this stuff, which is why IT usually has a change manager who does little other than checking this stuff.
Unfortunately in my twenty odd years experience of IT departments (as a non-IT employee with a clue about IT) less than 3% of IT departments and IT people I have experienced would have a clue about the perfectly reasonable stuff you wrote
Obviously!
If your working in a large enviroment that has implemented ITIL, then out of an IT department with 25 staff it's likely that a mere handful are actually IT people. Consider an ITIL IT Department:-
10 first line staff. Probably paid about ~£12k, (geeks with mild interest in IT if your lucky.) Capable of very basic troubleshooting via script eg. asking if the computer is turned on and logging everything that comes in. Has a hard limit of 5 minutes to a call and hated by people with real problems, but as they close > 60% of calls into the department management cares little what the users think.
5 second line staff. Probably paid ~£18k, marginally competent semi trained monkeys who can visit the user and plug the PC back in. (because the user unplugged it to put a phone charger in) or deal with other basic issues. Probably allowed to spend 15 mins troubleshooting, otherwise replace the PC and bring the box back to base to be reimaged if it's a software fault or have a vendor support call logged if it's a hardware fault.
2 third line staff. Probably paid £30k-50k+, real IT people who are ex-directory on the internal phone lists etc. These people deal with real problems and the other staff are basically employed to keep users from disturbing them.
8 Management. (1st line servicedesk manager, manager for 2nd line staff, manager for 3rd line staff, one real IT technical manager who used to be a tech in the days of token ring networks, process change manager, manager to manage the other managers, deputy head of IT, Head of IT.
Out of those 8 management one is technical, the others are process management, supervisory, planning, makework and irrelevance provided from the wider business.
So, in that 25 person IT department you might have 3 "real" IT staff. The others are just little cogs working to a narrowly defined process. In light of this, how surprised are you that you never speak to "real" IT staff?
We have entire leigons of minions (those first and second line staff) employed for no other reason but to protect us from needlessly unproductive interaction with users, because basically the business can't afford our time spent chatting to users.
Instead, you get chinese whispers from people assigned as go betweens who don't understand what they are doing, or why. Therefore, in the way of the world they try and cover up not knowing with waffle.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
