The rumours about Windows possibly being ported to ARM has left a lot of people bewildered; why would you port Windows NT when Windows CE 6.0 is a perfectly capable operating system? Putting all the pieces together, it’s actually quite clear why you would want Windows NT on ARM: servers.
The original report from Bloomberg focussed its attention on tablets and phones – in today’s tablet-crazy tech media landscape, that’s hardly surprising. Smartphones and tablets are very hip right now, and as such, it’s what people tend to focus on.
However, it makes zero sense for Windows NT to be ported to ARM so it can power phones and tablets. Microsoft already has the very modern and capable Windows CE 6.0 to do so; it’s the operating system, powering the Zune OS and Windows Phone 7. The latter, of course, is what most likely will become Microsoft’s tablet operating system – it’s designed for touch, after all.
It’s an additional story about this subject which gives a very important clue as to why Windows on ARM would make sense. The Wall Street Journal states that the release of such an ARM-powered Windows NT is about two years away – which left consumer-flashy-gadgets-zomg-sparkles oriented folk like John Gruber confused.
It doesn’t leave us OSNewsians – the more nitty-gritty type of geeks – confused, of course. About two years from now? That would coincide with the release of Windows 8, which makes sense when you think about it; that way, the developers can take ARM into account when developing Windows 8. If they were to port the current Windows 7 codebase alongside the development of Windows 8, the ARM team would have to do the porting all over again when Windows 8 hits the streets. This could be avoided by making the porting process to ARM a major bullet point in the Windows 8 development process.
But why would you want Windows NT on ARM? Simple: servers. ARM is currently developing very cool multicore chips for servers (virtualisation built-in), which could, potentially, revolutionise the server world. I’d much rather have a few passively cooled and very efficient ARM processors in my server than overpowered, energy-sucking, incredibly hot x86 chips which need complicated cooling systems to operate.
On top of that, Windows servers run a much narrower array of software than desktops, most of which is either owned by Microsoft (and thus easily ported by Microsoft itself), or open source (and thus most likely already available for ARM anyway). As such, the application problem doesn’t exist.
I’m fairly sure that if this ARM-Windows rumour is indeed true – and according to additional sources from Ars, it is – then we’re most likely talking about Windows Server for ARM, most likely as complete packages. I can also imagine Windows Home Server using such an ARM version of Windows Server, since it could mean cheaper, smaller, more efficient devices.
If Intel can foll around with Linux, why Microsoft can’t do the same with ARM?
Microsoft don’t like the vendor lock-in state. At least, not for them.
the whole thing is very simple really. remember the talk of having “windows as a service”? the OS loading from the cloud and bla bla bla. thinking about these arm netbooks and arm tablets it would make sense that these would load a web version of windows with minimal local instal. that’s a lot less to port to arm. then again, maybe I’m wrong. it happens every now and then.
You know why they want that?
So they can force everyone to pay a subscription to use Windows. The product on the computer itself would just be a loader to grab Windows from the ‘cloud’ (what we call the internet), making it virtually impossible to pirate Windows.
I can’t wait for this day to happen. It will spell absolute doom for Microsoft’s bottom line – in time.
The nerds will just switch over to Linux, or stick with old, pirated, Windows 7/8/X, and software will become increasingly scarce on Windows ( well, in relative terms only… ), and more plentiful in Linux et. al..
The end result is a larger, healthier, and free, alternative software solution. I love when companies fail to realize that their success is almost 100% the result of ‘undesired’ behavior, then fight that behavior only to see their doom…
Gonna be a few years, sadly… I hope they do it!
–The loon
Oh, and, what happens to all those wonderful Chinese? Oh yeah… they’ll just compete. Failing to obtain a vendor lock-in for that market is precisely what we need to fix our little problem…
EDIT: missing ‘r’, augmentation
Edited 2010-12-24 03:51 UTC
Another point on software, is that lots of applications for Windows are written in .NET, and would be easily ported to ARM, when .NET runs on ARM.
Actually .NET is not Windows. .NET applications are not Windows applications either. They are supposed to run on linux with Mono as well. Windows is not needed to run .NET applications.
Edited 2010-12-24 06:43 UTC
Try running Paint.net in Linux then.
Mono is a limited implementation and basically capped at version 2.
Windows ARM is a reaction to Google Chrome… and to a somewhat lesser extent the ipad… it’s not (only) about the servers… though I would not be surprised if the move was initiated by the Axure team.
The smartbook/netbook market segment is crap because there isn’t a great OS for them. Chrome is not impressive yet(give it a couple years), but it has to scare Microsoft. Windows ARM is likely to be their cloud OS for tablets and smart/netbooks.
Edited 2010-12-23 20:11 UTC
It doesn’t make any sense. Windows rack mounted servers ( the kind that would benefit from reduced energy consumption and heat output) are typically used by larger enterprises. Larger enterprises that *have* written software for windows server, and would have to port it over to ARM Windows.
PLus, I’ve heard the ARM server rumours as well, but I don’t believe they will be a viable option. There is today, not a market for ARM based servers. If you are moving from the mobile space, it makes a heck of a lot more sense to gradually move up in processing power: Ie desktops/net tops. Even tablets make more sense: they exist already and use ARM predominately. Microsoft has been abundantly ( and stupidly) suggesting over and over that their tablets will be full windows 7, not CE, not zune OS, not windows phone 7 OS.
Basically, I think anyone suggesting CE on tablets is just giving Microsoft too much credit to make a good decision.
“Larger enterprises that *have* written software for windows server, and would have to port it over to ARM Windows. ”
A lot of that is probably done in .Net by now…
Larger enterprises that *have* written software for windows server, and would have to port it over to ARM Windows.
In addition to the .net ancle already mentioned, “porting” means just recompiling. That may not be trivial for ad hoc hackjobs abundant in microsoft shops, but big hitters (oracle, ms itself, adobe, ibm…) will have their software up and running in no time (most of the time going for verification).
I wouldn’t be so sure about adobe, they can be a bit slow on the use of new platforms.
[ ] You have already done that before and know what a tedious and difficult job it can be.
Let me remind you, that Adobe and Sun took years to properly port their plugins to AMD64 which is “just” another platform in your humble opinion. Way more software nowadays still uses handoptimized code and leverages features which are only present on certain architectures making the code very difficult to port. Ask yourself why there is no native version of Skype for AMD64 yet and why the JIT compiler of Chrome and other browsers were available on x86 first.
Porting code to another architecture by just recompiling the code only works if you wrote your code with portability in mind right from the beginning. Unfortunately that’s not the case for many popular closed source software applications, especially when the codebase is older than 10 years.
Naturally, FOSS projects usually are way more portable since they were already designed with portability in mind.
And no, Microsoft certainly can’t just easily recompile the whole NT codebase, which is already 20 years old, easily for ARM. Hell, ARM is a vast different architecture, it differs much more from x86 than AMD64 even though the latter also isn’t just an x86 with doubled data and bus sizes (which isn’t even true since all current AMD64-compatible CPUs feature 48bit addressing only).
ARM is for example big endian as compared to Intel’s little endian and it’s a RISC architecture as compared to the CISC one of x86 CPUs.
Adrian
Really arm is strange. Arm process 6 or latter can operate in either big endian or little endian. What catches you out here. Symbian runs in little endian on Arm chips. So there is no endian issue using arm over intel.
RISC CISC is also not a major issue. Thinking Windows 2000 was on Power PC. So a few generation back.
Even better reactos a open source clone of NT already has a arm port that is fairly operational.
NT ported is not a issue. Applications for the ported OS is.
The irony here (and keep in mind, I’m a big FOSS dude) is that if I understand it right, NT is more portable than ReactOS.
FOSS != “designed with portability in mind”.
FOSS == “the code is there to be made portable by anyone who wants to port it”.
Linux wasn’t designed to be portable, originally. Read the original announcement email. It originally had a ton of 386 asm and depended on 386-specific instructions.
FOSS isn’t necessarily ‘portable’ code, it’s ‘able to be ported’ code.
Semantics, but yeah.
Reactos might be more portable than MS NT due to a lower amount of asm used. Please note Reactos is a NT design OS just built in the Open Source world from documents got on NT design.
Older the project in the FOSS world the more portable it becomes. Its a nature of the FOSS world as the backgrounds of the coders become more diverse so does the project supported platforms.
Yes a lot of FOSS projects started out platform locked one way or another.
Now closed source normally goes the other way. Portable and becomes less portable. NT use to support stack loads of different platforms but 2008 and Windows 7 support a small number due to cost cutting.
People forget that Internet explorer use to run native on Unix OS’s.
Look, NT _IS_ more portable, because it _already runs_ on multiple architectures.
Hell NT runs on chips not even made any more.
Will ReactOS _become_ more portable? Possibly. We can’t see NT’s code, or try to compile it on some obscure architecture, so if it becomes as portable as NT has been shown to be, we can assume a tie.
FOSS doesn’t necessarily become more portable over time.
That’s a fallacy.
It’s _often_ true, perhaps even _usually_, but _pleaaaaase_ don’t make blanket statements in favour of FOSS that _aren’t true_. It just becomes fodder for trolls.
ZSNES has been around for 10 years, and only runs on IA32.
Why? Because it’s written in asm.
BareMetal OS isn’t going to become portable because it’s not a priority, it’s not even close.
“Often” “Usually” but not “”.
Qualify the statement and I’ll say it’s true.
My simple point here we don’t know. Prototypes of Reactos exist for all arch NT use to support plus a few extras. Yes Reactos has be prototypes exist for those chips that are not made any more.
Reactos may be more portable than NT now. Just no one has really checked fully. I also understand why most people would not be interested to find out.
Older the project in the FOSS world the more portable it becomes. Its a nature of the FOSS world as the backgrounds of the coders become more diverse so does the project supported platforms.
You really did not read this. I did not say portable only in the sense of processor chips alone. Also I did not say the speed the projects become more portable. I should have been more clear in this. The speed of FOSS projects become more portable is unique to each project. But it is happening.
Even ZSNES does obey the FOSS world of becoming more portable with age as community around it has got more diverse. Being written in pure asm x86 for dos has yes slowed its progress. These days it has ports on Windows and Linux as well. Also it now has a splattering of C code appearing in parts of the engine.
Of course the speed of somethings becoming portable across cpu types may take 40 to 50 years to never with the community around and the complexity involved. Even with that limitation portability between OS’s and other things will still increase.
No matter how much you say no it will not happen FOSS it does happen over time and always will as long as the project has a growing community with growing interests.
Basically a FOSS project getting more portable with time is a sign of health about the open source project. Reasons why FOSS project will not get more portable with age all point to unhealthy things that either lead to a project forked or project fading away.
BareMetal OS is too young to know where it will and up and its Community is still very compact. You have to remember when Linux was young it was only intended for x86 processes. Coders around it grew and that ended as they become more diverse. Also BareMetal OS is not 1 kernel its two. 1 written in asm and 1 written in C. Of course the C one would be simpler to port. Yes even BareMetal possibility of portability has increased from when it started as a asm only kernel.
Really all the examples you gave to back your case back mine.
I think you’re misunderstanding the words that you’re using. Some grammatical issues tell me it’s quite possible that English is not your first language, so that’s forgivable, but your hubris at not admitting that possibility to yourself is not. If it _is_ your first language, perhaps you should study it a bit more.
I’m responding what you said, not what you meant; because I don’t have access to your thoughts.
Older the project in the FOSS world the more portable it becomes.
ZSNES has not grown any more portable for years. The portable C code is still portable C code, the i386 asm code is still i386 asm code.
It’s not a perfect example, but it still holds true.
If what you _said_ were true, it would have continued to become more portable as time passed. It hasn’t. Many other projects are the same. If it’s not a priority for a project, FOSS or not, a thing will not be done. If it is, it might be, even if it requires a total re-write. This has nothing to do with FOSS.
BareMetal states what it will not do. “It’s too young” is a foolish thing to say, because _they state it will never be ported_.
The article is about processor architectures, and you were talking about processor architectures. Why should I assume you were talking about anything but processor architectures?
“If it is a priority and feasible for a given project it will become more portable over time” is… something a rhetorical tautology, and pointless thing to say. It’s the closest thing to a true version of what you said, though.
“we don’t know” is false. NT has already had fully-working versions for more architectures. It _may become_ more portable than NT in the future, but as it stands ReactOS kinda-sorta compiles for PPC, and is pre-alpha quality on i386. You said ReactOS is more portable, because it’s FOSS. Not the case. It _can become_ more portable, but so can NT. NT currently has a more portable codebase.
If it was pre-alpha across i386, AMD64, i860(890? I’ve forgotten), Power, PPC and IA64, you’d be correct in saying it’s more portable. That’s not the case.
I’m trying to catch you in an incorrect statement before MS fanboys come in and say “LIAR!!!!11”
Say what you mean, and stop being so reactionary, because otherwise you’re just asking for trouble.
English is my first language. Yes my grammar is strange. This is what happens when you think the cure to dyslexia is study all the rules of the english language.
You end up with a very complex form of written grammar with dyslexia adding a percentage of error.
Big catch is that I gained 100 percent comprehension of anything I read. Sorry my poor writing is a double sided sword. I gained reading ability lost writing ability since. Too high of reading ability means you no longer can see particular style errors in your writing since what you are producing obeys the rules.
Yes I am lazy on doing , I avoiding putting them in because if I do. I will do even more complex english and heaven help a normal person reading PHD exam level english. Yes perfectly valid rule following english just not readable by a normal street person.
If you pull up zsnes source between versions and compare the code alterations over time you will see that what I am saying about portability is true and you are lieing you head off. The splattering of portable C to get into its code base has entered over a long time each time slowly displacing some of the x86 only asm along the way.
The speed is very slow we are talking a few lines per year of displacement in the portable direction with zsnes.
Your claim is invalid that is not moving in the natural FOSS direction. Speed slow yes insanely slow yes. But the FOSS nature is still there. Its like claiming that a slow moving iceberg that you cannot see moving in your direction without careful measurements is not moving.
Also the part that is a worry for zsnes future is how slow this move in natural FOSS direction . It says that the project has other competitors that splitting the community so with the passing of time zsnes could become an open source project that fades out of existence.
Next sign of fading out of existence. When was the last release 3 years ago. Now its competitors. bsnes Snes9x Mess all had releases in 2010.
Yes its important to accept FOSS nature. That is natural for open source projects to have particular things happening. When they are not is a sign of problems to come. Any old project you most likely will find to back you case you will find the same thing. The old project is on a death march out of existence unless something changes.
FOSS does not just cease to exist like closed source has a habit of doing. FOSS fades out. Slower release cycles. Until one day the releases don’t come. Then time latter the web site disappears and the storage disappears leaving behind only a faint memory.
Do have have to pull 1991 Linux mail chatter that says the same thing. Linus never thought that Linux would be ported off X86 either.
When BareMetal was started there was never intended to be a C version either. They even said never to C version and that there would only be an ASM version because they would be BareMetal. They are a young project the word Never from them does not mean much. Particularly that they have already broken a Never statement by making a C version.
Why does the FOSS nature of getting more portable with age stop or not happen and why don’t you see a lot of old projects with active development that go against FOSS nature. Lack of diversity in coders is the prime cause of not following FOSS nature. Lack of diversity also means limited supply of coders for the project. This leads to slower development compared to FOSS competitors that are portable since they can access a bigger pool of coders. That normally leads to the FOSS project not following FOSS nature becoming less functional compared to the FOSS competitor projects that is following FOSS nature simply due to lack of coders to create the features. So the one not following FOSS nature at some point fades away and is forgotten it ever existed.
Yes many forces of the system makes the older the FOSS project the more platform supporting it will be. Including death to those projects that don’t.
Now the same is true with Baremetal statements. As there coder numbers grow they are become more accepting of a lot of ideas. Even ideas the first group of coders said out right would never happen. Future of Baremetal is too soon to decide and too soon to take any statement they have made as absolute fact.
It is pre-alpha across all bar 1. The big issue is a lot of those have not been recently tested. i386, amd64, PPC, Xen and arm are all regularly tested. PPC of Reactos at this stage should work on a pure Power. Lack of hardware to check with.
i860 and i890 both full under Power and PPC.
This is why I said at this stage we don’t know. If everything in reactos is functional then NT is currently 1 down. If 1 is broken NT might level at the moment. Ie PPC x86 AMD64 Arm and Xen are all known runners with Reactos at alpha level. Power might not be operational and is the one question mark.
Yes I do know my topic here. You don’t. At one point I could list all the project leads in reactos for each of the supported platforms and there current state of operation of each of them.
“Untested” != probably works.
Xen != processor architecture.
Thrashing around in contrarian fury == priceless.
Edited 2010-12-27 03:49 UTC
Do you want to keep on proving you are not researching for making a comment. TheGZeus
Power was tested with reactos 5 years ago. It did function then. Problem here same sets of compatibility the person is giving to NT also has not be tested recently. So if it works or not would fall under not tested. So my allowance for classing it still working would be equal.
Xen has interesting side effects for processors that the OS source code can operate on. Of course a person like you TheGZeus I can expect to be a big enough of a fool not to know it.
You know the IA-64 support that Reactos is missing. The code to paravirt to Xen sees no ASM code required to that point. 100 percent none required. So yes Reactos can build for Xen paravirt on IA-64 and PowerPC from current source base without any issues.
Yes non native support I did fairly count it for the extra platform that was not already supported. IA-64. I did not recount the PowerPC x86 or AMD64 support a second way through Xen.
Of course Xen support for a CPU is different from that of direct to hardware. You don’t need to init the CPU or setup lot of basics because the Xen Micro kernel does that.
Look, I understand dyslexia makes it hard for you to write.
Please understand that it also makes what you write impossible to read.
I cannot comment on the veracity of anything you’ve said here, because it makes neither grammatical nor logical sense.
I, myself have dysgraphia, though it mainly affects my ability to write by hand, and I have very odd typos when using qwerty (which I learned not by touch-typing from the first, but visually, so it’s all weird in my head).
I have empathy, but little sympathy.
I _think_ you said something like
“It worked before! Virtualisation complicates things! This all means I’m right and you’re stupid, so shut up! NYAH! I’m not listening LALALALALALA!”
Oh, and cloning a dead/dying OS to support legacy code on old hardware: Great idea! Have fun with that. I’ll be over here, away from the cruft.
Edited 2010-12-27 07:46 UTC
Write is simple. If you will be a level to read it that is another matter. I have gone back and check my percentage of error was low.
From my check of comments a total of 8 words wrong. Most likely no worse than most peoples typo rate. Yes when 24 hours have past I can see my dyslexia caused errors.
But my higher than average use of exotic english rules. I know can cause trouble.
Don’t give me this cop out. Quote the blocks you cannot read.
If you could not understand what I was writing. Why did you try to draw ideas from it. Not had the balls to speak up about your problem until now?
What you think I have ESP and can read your mind from most likely the other side of the world?
If you are commenting and you fail to mention that you cannot read something. I am in my right to presume you been able to read it.
Lot of logical sense problem I think is that you don’t understand the subject matter to my level. So you are unable to see why what I said is logical.
A cop out defense. Basically every point you are pulling out of mine has backing behind it. Instead of querying you have be insulting. I tolerate being queried.
I don’t tolerate being told hey you are completely wrong where I know for that fact I was not. This is exactly what you did. TheGZeus. And have been doing repeatedly to others as well. So its not just my english here.
You have proved over and over again you don’t really know the topic.
On the Xen one nicer would have been “Why are you counting that as a platform to be it does not appear to be. So I take the numbers as the follow as correct in my eyes.” But no you had to take the short but insulting path because you did not believe I could be right. So being disrespectful.
Of course the funny part here. The currently tested platforms of Reactos are chipsets still in production. Not legacy. Reason for pure Power not being tested is that chipsets using it don’t exist devices easy to test a OS with. Some NAS box’s use a SOC chip containing a Power core. Yes a old Power not a PowerPC.
Reactos is also cloning API from Windows 7. Again not legacy. To clone Windows 7 you need to clone large sections of NT.
Again how many times are you going to comment without doing you homework. Or will you change to being respectful and at least presume the other person has a brain and might be right and ask for clarification.
If you are going to make a statement that someone is wrong you better have the documents backing you.
IA86 is about the only one completely gone and it makes sense for a modern OS developed OS to support that through an item like Xen. Of course you could have found that out nicely by asking the right way and not presuming you new the topic.
Simple fact TheGZeus being disrespectful makes me respond nasty. Should I not treat you how you are treating me?
I don’t have to put up with people being insulting particularly when they don’t have the documentation or research behind them.
“On two concert I’m should collective photo, but such small fat bald-headed technologist be insane.”
The meanings of both of these can be discerned with effort. You’re not worth it.
I’m done. This isn’t productive, it’s no longer interesting, and you are obviously far more willing to pour energy into this.
…and it seems the more energy you put into it, the less sense you make.
Oh, and IBM still produces Power chips.
Did you say you’re on some next-level knowledge above mine? You didn’t even know what architectures on which Windows runs.
That’s a _guess_, and I was able to discern that much because I talk with people from Eastern Europe and Asia all the time. I also said that I couldn’t speak to the _veracity_ of your arguments. I didn’t say I couldn’t get the general idea.
_I’m_ beginning to doubt dyslexia at this point. Your grammatical errors follow the patterns of English-as-a-second language pretty well, though your error count _is_ very high in these past two posts.
It’s not a cop out, and you’re just being lazy if you can see the errors and not fix them.
Why would one want to use Windows instead of porting a program? Legacy proprietary software.
You’re what, 19? 22? You don’t know everything. You’ll figure that out in a few years. Don’t try to battle me with logic, because (once I figure out what the hell you’re talking about) you’re making leaps of logic, not following things through to the conclusion, and failing to see your own hypocrisy.
You’re either completely without wisdom, or a troll.
I’m out.
Edited 2010-12-27 15:02 UTC
Power the chip brand and Power OS design arch are different very different in meaning. Again leap of faith on your part that you have found where I am wrong.
PowerPC and Power3 on all will run PowerPC code in fact with PowerPC code on all this chips you can detect then run the correct instructions for the later power processors. In the same way you call all i386 i486 pent the name x86. The PowerPC and latter chips as a platform get called PPC or PowerPC. Yes I know I should have used PPC everywhere.
Power version 1 and 2 the early design is only in SOC form these days. Should have stated this these are hard to develop and debug OS unless you the maker who ordered the SoC on due to the fact each one is basically 100 percent custom. These SoC still using Power 1 and 2 due to dia size savings less instructions less silicon more room to fit other features.
Sorry Lack of logic on your part. There had to be a reason why I was referring to Power and PPC as two different things and why I said Power was only made in SoC form. If you had checked out the versions of Power only produced in soc currently form you would have found those are only Power 1 and 2. So showing the difference.
Claiming I have leaps of logic. The text I said was logical and provided all the logical information to match it up. Yes light on for information. You have been doing leaps presuming wrong again. Instead of asking
So leap of faith and hope you are right. You general idea is very much like being in the USA state washington instead of being in washington dc. Both are washington right? Stop doing this it will keep on getting you into trouble.
I am sorry to say your track record for getting 1 thing right from your general idea extraction. Makes my english error rate look minor. You would be better to ask for clarification then depend on your means to extract the general idea. Because you have failed todo this.
You know the funny part here. English is a diplomatic language. A lot of what people call grammatical errors follow the patterns of English as a second language. Are not errors. In fact part of permitted english syntax. The difference is normally a english person would only have 1 or 2 in a document. I am likely to have 1 or 2 every block of text.
Sad part is lot of people with English as a second language get berated for having poor english when there stuff is perfectly permitted. When they go to correct it they end up with poor english due to the fact they end up with english rules wrong. Basically the language equal to racist lot of the markers people use to detect if a person is using english as a second language. Even worse really. Thinking most of the markers used are commonly used just in a lower volume by normal native english writers. Lot of cases in the very material they used to learn English from. Since it suited there native syntax they take them on-board leading to there higher than normal rate of usage.
But higher than normal does not mean incorrect english. Just different english. Call it a tolerance issue.
Because of the lag its 24 hours before I can see them (Yes the annoying side effect of my form of dyslexia) and here does not allow editing of comments after other posts. Other forums I do fix them.
Also if I am pulled up inside that 24 hours It helps me. What I cannot see if someone points at it I can see it. This is why I am annoyed with you.
If you have a problem speak up. This way I can attempt to prevent it. You don’t speak-up there is nothing I can do. Knowing there is a problem is key.
I ask for you to pull a block for a reason. Once the block of text is out of context of what I wrote by someone else. My mind means to hide the error can disappear so allowing me to see other errors as well in the first thing I wrote and go opps.
Yes I also have a degeneration issue. Error stacked on error. Speaking up correctly stops this from happening. Simply you don’t want really bad to read from me see error speak up.
When my dyslexia gets completely out of control. there will not be a word in the right place that is completely not readable by anyone including me after 24 hours. If someone speaks up in 24 hours and I see it in time I can recover what I wrote.
So what you have seen is only the minor edge of my level error and you are having trouble and not speaking up. So I fear that I could end up at my worst.
Claim I am without wisdom.
Is NT master design bad. Answer NT master design is quite well thought out from a secuirty point of view. Implementation of NT has been lacking.
NT on arm has no Legacy proprietary software yet Reactos is there. Question why.
The shocking part to most people is that MS so called platform neutral .net lot of the applications made in it depend on the way NT designed os’s operate. Even some of the open source .net applications. So the design is not as Legacy as people like you think. As long as people keep on coding in .net using windows nt platform only ways there is a place for a NT clone. Yes the ways used are not even windows ce compatible.
Lack of information on the topic and it shows. TheGZeus. Yes I work with the wine project to allow legacy to work. Yes large numbers of .net applications are simply not platform neutral.
Edited 2010-12-27 23:55 UTC
The current Power servers have all the PPC instructions.
Everything you say proves you _wrong_.
At this point, I’m calling bullshit on dyslexia.
You’re just the average troll.
To be correct where did I say limiting to Power chips made as servers in production. Yes current Power servers are all PPC class for OS development. I never disputed this.
If Power 1/2 chips were in normal servers they would be simple to get to test with. Logic should have told you when I was saying hard to get to test with it would not be this simple.
This is your complete problem you not reading what I typed. Power SoC chips are not just found in what you call Power Servers.
Power 1/2 SoC chips made new today are found custom items like NAS hardware, some printers and some forms of routers. Not the general Power servers that are PPC class. Reason why testing for compatibility with current day hardware is hard. Not impossible of course but hard.
How is this being proven wrong. I told you the class you were looking for with Power 1/2 custom SoC chips. Do Power Servers contain custom SoC chips. Answer no. They may contain a SoC chip but it will be a mass produced form normally with standard socket allowing simple replacement in case of malfunction.
Basically now you are going to claim you could not read what I wrote. So your made mistake?
Now if you are incompetent to check in the wrong class of hardware and bring the wrong answer back. This is not my fault.
So its not been everything proving me wrong. It been someone not reading and doing invalid research trying to make arguments to win.
At the moment my dyslexia rate has been low. I am truthful about it. There have been a few word swaps from my dyslexia but lucky most of them were no worse than normal typos not complete context changes. Those errors are spotable if you look for them. First and last letter of word most cases matching intended word. Letters in middle incorrect so equaling different word. Yes there is a signature mark of dyslexia if you want to look for it in my posts. I do a very good job of controlling my dyslexia most of the time. Problem is I cannot depend on always being in control of it.
English complexity and edge rule usage been a little high.
Edited 2010-12-28 04:43 UTC
My first paraphrase is valid.
First: many PPC boards can be had for relatively inexpensive.
Second: Power servers are in production, not legacy.
You said (paraphrased from jibber-jabber) “yes they still make power servers, but custom soc machines are not Power 1/2, so shut up”
What I said in my paraphrase was
“It worked before! Virtualisation complicates things! This all means I’m right and you’re stupid, so shut up! NYAH! I’m not listening LALALALALALA!”
You then said that PPC was dead (XBox 360 anyone?), to which I said that IBM still produces them.
You then went on some rambling rant about how Power is different from PowerPC, and Power isn’t used in SoC boards, so I shouldn’t say that Power is in SoC, or something…
Email it to someone, use a Grammar Check tool (is there a FOSS grammar check? don’t remember, because I DON’T NEED ONE HAHAHAHAHAHA), or shut up. If you refuse all of this, and your error rate increases while you say things that would be completely impossible to make sense… Trollololololoolol
Very hard to understand to gibberish you’re putting out, (intentionally, I’m convinced at this point).
f–k
Off
“when you get aggression, give it 2 times back” -Joe Strummer
Edited 2010-12-28 13:26 UTC
This isn’t usenet or IRC. You can bold and italic the text.
On the topic:
Wasn’t the latest NT kernel a substantial rewrite? I think it was…
NT once ran on systems that are no more, so it’s no longer true. And having over 5 years of development concentrated on x86 brings in a lot of non-portable code.
We can argue as much as we like, but the fact is that we just don’t know. Unless you happen to work for MS in that division.
JAlexoid,
Windows 7/Server 2008 R2 also runs on Itanium. It’s not x86-specific.
Microsoft just never released a workstation version for Itanium after Windows XP because they only sold about 300 licenses.
Exactly.
Is it so hard for some people to hop on Wikipedia?
To paraphrase you:
Yes, because posting a comment gets the same info with so much less effort than going to Wikipedia.
Yes, because 7 characters gets the same point across with so much less effort than two.
O_O
Not much changed with server 2003, and 2008 officially runs on Itanium, so my point stands. The original statement is false, the new one is moot.
We know the latest server products runs on three architectures, and ReactOS works its best on one, and kinda boots on another.
You’re _wrong_. It’s not bad to be wrong. It’s bad to refuse to admit it in the face of irrefutable evidence.
Edited 2010-12-27 03:06 UTC
I work on NT. Just like any other OS (Linux, BSD, etc), it doesn’t get rewritten every release. Some code gets changed, other code stays the same. It’s more evolutionary than you seem to think.
It’s quite disappointing to hear that major version change did not yield substantial changes in the kernel…
It’s not a cost-cutting issue, those platforms just did not sell at all or had really bad marketing.
Compaq killed the Alpha right before Windows 2000, despite the fact that it was the most powerful platform at the time to run Exchange or SQL Server on. I saw 64-bit Windows 2000 running 64-bit SQL Server 2000 at PC Expo in 1999 at the Javits Center. It smoked everything x86 at the time. However, Compaq must have learned their marketing lessons from Commodore because it was never publicly released, and soon after, the Alpha line was killed off. There was genuine demand for their products, but they managed to screw it up.
The MIPS and PowerPC platforms also did not sell at all. They didn’t sell much with anything other than UNIX or Mac OS on them. They both got killed after Windows NT 4.0 Service Pack 3. There was no way Steve Jobs was going to allow Windows to run on a Mac right after the clones (now is a different story!), which took away 90%+ of the PPC machines out there, and since this was soon after Microsoft gave the finger to IBM with OS/2, no way you were going to have it running on their POWER HW.
The Itanium managed to stick around until Windows Server 2008 R2. After this one, it’s gone, and the only big-name OSes you will be able to get for an Itanic will be Linux, OpenVMS, NetBSD, and HP-UX. I’m sure there still will be ports of various other OSes, but it’ll be pretty much dead outside of HP.
Microsoft is the kind of company that can afford to have multiple “Project Marklar” operations running. I’m sure that they have Windows running on ARM, MIPS (the Chinese variant), IBM POWER, and several other architectures (I’m guessing SPARC64).
Yes there is. It’s not going to replace x86-64 overnight, but Linux deployments make up a significant portion of server sales. Linux deployments mostly don’t need to worry about the CPU architecture.
I know I could replace a significant number of x86-64 servers with ARM without any problems, and the power savings are likely enough to make it worthwhile. I don’t believe my situation is at all unique.
What do you mean by “Without any problems”? You haven’t benchmarked any of the ARM servers for your load, because they don’t exist. How do you know these mythical servers will be able to handle the current load of your x86-64 servers?
I was talking from a provisioning & operational standpoint. Whether a particular server can handle the load you intend for it applies to any CPU, no matter if it’s ARM or x86-64.
If it helps, the majority of my x86-64 servers are virtual machines with very low CPU usage profiles: I have some physical hosts running multiple KVM instances where the Munin graph for CPU usage is basically non-existent.
Ok, that makes more sense. But within the same processor architecture, its pretty easy to just guess that this years x86 processor is probably going to handle the load that last years did. Changing architectures means you have a lot less in common and requires much more intensive resource planing.
Virtualization, in my opinion is a good argument why ARM will never take off in the data center. If you have a lot of underused servers, virtualization is awesome. A much better solution for high availability, backup, and energy conservation than ARM.
Counter-point… One of the largest growth sectors in the server hardware market is large data centers for cloud computing. What you want in such data centers is small, cheap, low power (critical for operational costs) servers that generally do one of two things:
1. Run custom written distributed database/file storage software.
2. Run custom written web servers/caching proxies that are purpose built for cloud applications.
Linux practically owns this market as it is now, and ARM servers are more than likely going to be very attractive assuming they offer advantages for power use/thermals/size/etc. Since Linux is already the defacto standard, ARM has no real obstacles in its way to enter that market, so it is likely to at least get its foot in the door. Microsoft would like a piece of it. Simple really, I frankly don’t see how they could ignore it.
Porting Windows (in one form or another, I’m not claiming it will be exactly the same as on x86, but CE is simply too striped down for this type of use) to ARM would in effect give them somewhat equal footing once the hardware becomes attractive enough that customers start buying it. True there is something of a challenge in porting the software required for these types of deployments, but it isn’t inconceivable that Microsoft could offer some advantages here that might sway some customers (i.e. existing Microsoft shops wanting to build purpose built cloud services).
Point is without any presence on ARM – if ARM servers take off Microsoft is completely locked out of a potentially huge growth market. Do you really think they are willing to let that happen?
This was my take on it as well. New power-efficient ARM chips that could be used in a desktop or server role are on the horizon, and they will be available to market in a couple of years. Some OEMs are sure to take advantage of such chips and make a server which achieves a significantly better power-per-performance metric than x86 or x86_64 can, perhaps it will be more than an order of magnitude better.
Currently, the only viable server OS for such a machine would be Linux.
IMO, Microsoft cannot tolerate that. If Microsoft want to stay in the server business in a couple of years time when these machines come to market, Microsoft will HAVE to have a Windows server on ARM product ready to roll. If they are even a little tardy, Linux will take over this entire market segment.
If you read this article http://www.linuxfordevices.com/c/a/News/ARM-CortexA15-MPCore/ , it explains the larger reason’s for the future demand for this. Energy savings, and virtualization.
First, even the appearance of being a “green” company could bring large business your way. Think of Google’s effort’s over the last few years to cut down on their carbon footprint. It’s saving money ,also, as well as the environment . Then think of the virtualization solutions on big iron servers, where the base operating system ( Linux, most likely) host’s any other operating system that is required . Take a look at the top 500 fastest super computers to see how many aren’t Linux, and you’ll see why this could be very important to Windows. IBM was king of the big iron at one time, a mistake Microsoft doesn’t want to see happen to it either. Also, imagine the sync capabilities of your tablet, phone, home and server being all one type of chip and Operating system. Much more at stake than just iPad’s and iPhone’s. Just my two cents.
I do not agree, as an Admin of such a company I see the majority of these servers run IIS for asp/aspx, MS SQLServer or Java.
All that stuff will easily run on ARM.
Many Webservers out there only act as proxies providing SSO for WebLogic and other backends.
These webservers would run just fine on a dual core ARM with 2GB RAM.
For DB an application and database servers you still will want maximum CPU and memory.
Currently servers is the only place Microsoft can comfortably try new things. Most of their stuff is based on .Net, which is portable anyways, and the NT kernel has always been multi-platfrom capable.
And, if they succeed, I believe, like others mentioned above, this will pave roads for a future Chrome OS competitor in smaller devices. If they can make sure that Windows runs OK on ARM, then they can have a small shell/IE/Silverlight/RDP services for their lightweight clients.
Most of their stuff is based on .Net
I assume you were not thinking about Office, IE, Active Directory and other similar behemoths. They are most likely in C/C++ with some portion specific to processors(Assembly!), porting them is little more than just compilation.
Most of their tools are nowadays developed in .Net.
Even the C/C++ build chain now make use of msbuild (Microsoft’s Ant) instead of nmake.
The only C/C++ software are operating system low level stuff and APIs, games and applications that Microsoft has before .Net.
Still, C and C++ can be made quite portable, and lets not forget that the NT line used to run in several platforms.
Thom, now I know you want this site to be more about discussion of OS related news, but this looks like your pure speculation with no real hard evidence.
one of osnews missions statements is to explore to future of computing. Discussing the future inevitably involve speculation. The speculation here is not wildly unrealistic and the articles and many comments involve educated guesses that make for good reading.
You will find absolute no news site that dont have speculation articles among its news. Thats true even for Reuters, cnn and others.
Sometimes even discussing the future actually shape the future.
But that other speculation usually has at least one verified source of the information being used, this article’s only real source is an article that is also pure speculation. No expert, no insider, no Apple factory worker…
But this is CES, the Consumer Electronics Show ! You are not supposed to talk about server stuff at the Consumer Electronics Show !
So post example for 10 different arm based rack mountable server, I can’t wait to read about them also post what os they come preinstalled with and on what price.
I never ever come across arm based servers, only embedded devices. This speculation is so stupid like fbi putting backdoors into openbsd. I think you should delete this article out and write one about vmware vs virtualbox performance tuning or something useful
Your assumption of ARM is most definitely wrong. Take a little loom back into history, and you will find that server CPU architectures change from larger, more powerful chips to smaller, more energy efficient ones. Just look at how x86 took over from Alpha, PowerPC, all the other larger chips, reducing cabinet size systems to today’s blades. ARM servers are coming, and there is much proof of this. Just the other day I was reading about a new ARM 4U server with something like 80 cores in it, and it took less power than a traditional 1U server. That kind of density is impossible with even quad core blades. Here’s a little light reading:
http://thinq.co.uk/2010/11/30/zt-systems-launches-16-core-arm-serve…
http://gigaom.com/2010/04/28/cell-phone-chip-king-confirms-its-serv…
http://www.theregister.co.uk/2010/11/02/dell_dcs_arm_risc_server/
http://www.linuxfordevices.com/c/a/News/ZT-Systems-R1801e-/
As today’s ratings change to performance per watt, ARM will make great leaps into server rooms already struggling with cooling issues. Will Intel be defeated? Of course not. Intel chips will always have a place in scientific computer and other very processor intensive tasks where optimized kernel libraries and hardware tricks can accelerate code execution, but for everyday database servers, web servers, and the like, ARM will be the future, a future that takes less power.
This is indeed the first arm server i see. GUESS Cant buy one for home yet cause it cost like 20k usd.
“Data centers spend billions on electricity for servers, including energy used for cooling, and they have a significant CO2 emission footprint,”
oh please… all they care about cutting costs because they can profit more on it. WOnt mean that you as customer will get server from the resellers cheaper. Sad that even with the widespread use of virtalization products most of the servers are idling in data centers. I think if we would summarize the idle time of the servers in all bigger US data centers it would be like 70%. And besides this the only bigger waste what us military and agencies do, building bigass clusters for ripping their pron dvds which also mostly idling and wasting energy so keep qqing about energy consumption.
Edited 2010-12-24 19:11 UTC
…and if things go well with regulation, COâ‚‚ emissions will be taxed to hell, thus causing them to cut into their profits.
Edited 2010-12-25 01:39 UTC
At that very CES2011 show a windows 8 tablet is going to be demoed also. What a double wham it would be if were going to see a Windows 8 ARM tablet.
I doubt whether they are going to use NT though.
Why I think that is that if Windows 8 is going to have that fabled new filesystem, arm support might have been build into it from the very start.
http://www.engadget.com/2010/12/13/microsoft-to-demo-new-slate-pcs-…
http://www.fudzilla.com/notebooks/item/21312-windows-8-to-run-on-ar…
sorry
Microsoft doesn’t have anything in the SOHO market that supports Active Directory or cloud. They don’t have that unifying product that can stave off Cisco or their Linux competitors that makes running SOHO IT easy and integrates cloud services.
Think of a NAS appliance that runs AD, Forefront (new ISA Server), SQL Express, WSUS, App-V, a Microsoft Security Essentials server, and connects to Microsoft’s cloud services to run hosted Exchange, Sharepoint, and Azure. Think of being able to run third-party tools and services such as WebSense on this box as well.
Think of them being able to charge a monthly fee for data backup, services, and upkeep. This would create a cloud revenue stream that they do not currently have, provide a seamless interface for backing up AD and doing lots of other tasks companies require sysadmins for, and remove the need for dedicated admins for Exchange and Sharepoint from small companies.
Think of the fact that SOHO PCs can connect to this, be autoconfigured for WSUS, Exchange, Sharepoint, App-V deployment, Microsoft Security Essentials AV/Anti-Spyware, and web filtering, and work.
It’s a very good strategy for companies that don’t want to run a complex server infrastructure to get these features, and don’t want to pay the huge initial licensing . You put a little 1U box the size of a network switch next to your Internet gateway, run a few wizards to configure it, and have Internet, content filtering, backups of all of your data, and the ability to restore it from a web browser.
Additionally, they can bring in selected third-party partners to run apps for it, put an App Store in for applications (with the various charges), and make running AD and Windows dead simple. They can also provide this as an endpoint for distributing applications such as Office via App-V.
Like I said, Cisco is getting close, and Google is offering cloud services that don’t encompass everything the way Microsoft wants to.
Microsoft has a lot of tools at their disposal. If they can pull it together and offer their technologies tied together in one neat package like this, they really could present a valid alternative to companies looking to make network administration painless, and establish a recurring revenue stream to make money. It would also re-allocate a lot of money that goes to the likes of Symantec, McAfee/Intel, Carbonite, and Google to Microsoft.
Most importantly, it means a recurring revenue stream that could be worth billions of dollars a year that currently is going to their competitors and partners. Unfortunately, it would give Microsoft a degree of control over PCs that they only wish they could have had previously, since they would have the ability to turn off the switch if someone doesn’t pay the bill.
Windows Server is probably already running on ARM anyway. Microsoft doesn’t release all of their ports of Windows NT (remember they did port to SPARC, and reportedly to the 68K processors). It would not cost much money to build these devices and sell them at an initial loss, while making up the cost on a monthly “utility” bill.
And here is the shocking part. Linux is inside striking distance on that complete list.
When samba 4 releases almost straight up Linux will have Exchange replacement. Sharepoint vs alfresso Sharepoint normally loses. Virtual apps there are a lot of those for Linux already.
MS is basically forced to port to arm or be broadside not far into the future.
And the only vendor who’s been able to capitalize on Linux in this market and make a decent profit has been Cisco with their Linksys line of products. Granted, there are a ton of other smaller solutions out there, and quite a few Linux distributions, but they’re never going to get the penetration that Cisco will.
Canonical or Red Hat can pull this off with the right subscription model and OEM/partner support (Dell or another enterprise vendor, since HP is too committed to MS). Amazon, Citrix, Cisco, VMWare, Apple, or Google could also pull this off. At this point only Cisco owns the stack and enough of the technologies to bring this to market.
For what people think of Microsoft and Sharepoint, they have a ton of applications and technologies that integrate into them very well that they own. Sharepoint has very robust integration services that plug in everywhere, and provide services that the other Intranet services don’t have. This is their best-selling new product in years, and they are converting many proprietary apps to use it as a back-end (Project Server, Exchange Public Folders, etc.). Open Source will not replace 100% of that integration. It works fine for many smaller projects, but not for Excel Services or Project Server.
Microsoft has a very large portfolio of products that they have never integrated well, and all signs point to the top. If they have actually gotten an executive into place that realizes what the customers want and need, and figured out that they have many hardware partners that can push out these products cheaply and quickly (Dell and HP come to mind) while integrating several product lines and changing the business model to focus on recurring payments (which is what a CEO is supposed to do!), then they have a large chance of success with Windows 8 Server on ARM as an underpinning to a cloud solution.
And you can bet that when Microsoft starts selling cloud, they will be hammering Google to the wall for their privacy violations, and talking about how Microsoft meets every federal/EU standard with their product lines. They will be talking Green Computing, and they definitely will be talking about cutting IT costs and making PCs more manageable without having to integrate several products. They will also talk backups.
The big thing you can hope for is for Steve Ballmer to shoot himself in the foot. At the rate he’s going, it’s more likely to happen than not. I think most people realize that you could put any half-decent IT executive in his place and they would have done something similar already.
You really have zero understanding how close Linux guys are. Exchange Public Folders don’t make me laugh. sogo using openchange does that as a drop in replacement.
Alfresso has Project Server features built in. Also Alfresso pretends to be sharepoint quite well.
Samba 4 gives drop in replacement to ADS.
All these don’t have to use CAL’s.
The delay is Samba 4 its a keystone. MS has to move before the Keystone becomes functional or they are going to be hit by a boardside of matching techs on a platform that they don’t have a OS for that is lower cost than the platform MS runs on.
Sorry claim that opensource will not replace 100 percent it basically ready to replace at least 90 percent. What for most businesses will be good enough. Since it will replace the missing 10 percent with other ways of getting the same results.
Arm and open source on servers threatens to bring the 500 dollar server todo most business needs into existence with unlimited cals.
The Linux guys are always infinitely close – but never quite there.
while in other aspects, windows guys are infinitely far and will never be even close. instead of wasting money on generic solutions one can order tailored ones for that money and get way more bang for same amount
all depends on what one needs. if i’d be moving me or any of my customers on windows i’d be crippling them with added additional costs. all software they need and services are already tailored to linux primarily and there simply is no simple and cost effective way to introduce windows and generic solutions you name. (i just introduced this aspect from my point of view to produce contra to your personal aspect. I’m not even nearly bashing ms products. if anything… i say “use the hammer that works for your nails”)
Not true. It takes a long time to get all the interlinked parts work work. Did not help that a Legal case had to be won and MS made obey to get the information to get this close.
Illegal actions is market kept MS possition. Linux was doing quite well even gaining before 2000 and MS illegal action. Basically the same thing that shot Novell shot Linux. Not been something simple to recovery from.
No, I do have full understanding of how both work. Sharepoint is about the only thing Microsoft got right in the past 10 years besides Windows 7.
Linux has been close for years. I remember when people were selling Samba 2 solutions for this years ago (remember Cobalt, who Sun bought? I was running them in 1999).
Alfresco pretends to be SP quite well, which in itself was Microsoft’s reaction to Oracle’s Content Management Server (which started as Internet File System in 1998-1999). However, there are features of Sharepoint that Microsoft pushes that many of the large vendors such as SAP, Kronos, and a very large amount of software vendors use. Your average C-level with purchase-level authority doesn’t want to hear “like sharepoint”. Project Server in itself is a pretty complex project that sits on top of it as a front end. For every Microsoft product that uses it, there’s another 10 industry-specific solutions that do as well.
I’ve spoken with Microsoft many times, especially about their latest push to build applications on top of Sharepoint by hitting the verticals. They are aiming dead at Adobe, EMC, and Oracle with this, who have solutions so expensive for these that they will make you wish for CALs. Those who complain about CALs have never had to license Oracle or Adobe LiveCycle.
Exchange still has a large amount of the marketplace. I gave that as an example as how Microsoft is transitioning older functionality to newer products.
BTW, I wouldn’t run OpenExchange when Google is right there with better functionality and the ability to plug into your LDAP/AD system.
You have no idea of what I was aiming at. Microsoft is trying to replace the idea of CALs, which are the devil, with a subscription-based revenue stream. They actually have all of the technology to put something together that customers can use and is very simple to use. They actually can package it together and sell it. Cisco, via their acquisition of Linksys, also has this knowledge in-house, and has sold products that have this functionality. Oracle, Symantec, Apple, and Google also have the technology, and have implemented it in various forms.
Microsoft is losing money to IBM, Google, Apple, Symantec, Open Source, Cisco, and Intel/McAfee. They have the ability and technology to integrate everything and present a solution to the customers that is a one-box solution.
Canonical (especially since they do a lot of ARM development) or Red Hat, if they had the inclination, could do exactly the same thing with the right hardware and software partners. Novell got bought and neutered by Attachmate/Microsoft and is no longer a threat (and they have the knowledge in-house). They were the Commodore of Linux.
This isn’t about having a solution, it’s about having an integrated one.
^
This.
This is exactly where our business is moving. We have SAP and other systems that Sharepoint can integrate with … it is a no brainer for our business since we already run these systems and sharepoint can integrate with these systems quite easily.
Also because we are using Sharepoint elsewhere in the business we are rolling it out to our website, mainly because we will have a lot of in house skills with sharepoint at the time.
Edited 2010-12-24 20:59 UTC
Yes 1999 Linux was doing quite well. 2000 MS did a illegal action and that was the end of Novell netware and Samba from domination and MS took over.
Court cast had to be fort that was won in 2006 and its taken quite a few years to make up for 6 years of lacking information. The not quite there state being as bad as it is results of an illegal action.
Alfresso I said pretends to be sharepoint. This is API level. So to SAP it thinks it is talking to Sharepoint but its really talking to Alfresso. Anyone who looks at the user interface knows Alfresso looks different ans has other features that Sharepoint lacks. Ie pretends not like. I have not tried with Kronos but I would not be surprised if it worked as well with Alfresso.
Of course they are trying to build vertical on top of Sharepoint to lock you back in.
They are about to lose the vertical of Exchange and active directory that allowed them to destroy Linux and Novel netware from 2000 on with illegally closed protocols. So they have to create something new.
Openchange is not openexchange. openchange can also be used to back end onto google. So you don’t need to install google sync on client machines. Basically openchange is a wrapper allowing outlook to see an exchange server but many different groupware servers be hidding behind it. Since it is being a exchange server not a installed add on it cannot be damaged by MS doing updates to outlook unless MS wants to break compatibility with there own servers.
And there are many Linux distributions that are very close to having integrated really to go. Not the players you are listing either.
http://www.zentyal.org/ http://www.clearfoundation.com just to name two.
The keystone landing I am talking about will release new players into the market on equal footing to MS in lots of ways. Past MS in another.
This is exactly why I come here, to learn about great new stuff like this. I’d been mucking around with FreeNAS trying to add webhosting, and email serving to is manually. Both of these appear to be exactly what I was looking for + a lot more.
Thanks for the heads up!
Ok, good business pitch for a new windows server/ app bundle. What part of that requires or would benefit from a ARM processor vs x86?
The fact that they can pull a reference design and quickly get this to market without having to do much engineering work.
Additionally, the fact that they can use this to set up a closed infrastructure where they can let people into their “walled garden”.
Finally, the fact that you can shoehorn this into something with the dimensions of a network switch due to ARM taking a lot less power, and make an appliance out of it. Atom just isn’t there yet.
Can’t say I agree. X86 is not lacking for engineering designs. Even if it were, would that be more work than redoing all of the software you mentioned in an ARM compatible manner? I seriously doubt that.
Plus, remember, your target audience is a Small or Home Office. Power, space, heat: these are not that important in this environment. The power savings would be minimal for a single device.
I mean, look at this for heaven sakes:
http://www.msi.com/index.php?func=prodtmpspec&maincat_no=729&cat2_n…
35 watts. That’s a compact florescent bulb.
Now, I understand the Ipad (a powerful Arm device) can pull as little as 2.5 watts. A significant difference when running a large number of devices.
At an average price of 12 cents per kilowatt hour here in the US, that amounts to yearly costs of
MSI Wind : $35
iPad : $5.00
So they’ll be saving every company $30 on average by porting everything over to ARM, assuming that they are both capable of handling the required work load.
35 watts ??? For an x86 PC with these specs, that’s HUGE !
I mean, there’s much more than an Atom and an intel chip in my laptop, and still it consumes about half less under normal use… And it’s a known fact that these mini PCs are made of laptop components.
Maybe they were talking about peak power consumption, though.
Edited 2010-12-24 09:59 UTC
Server room numbers are peak and complete including duel harddrives both spinning explaining about 20 watts alone. So about 15 watts for the core so that is a very power effect.
Yes the raid in the mess ruins you day.
Yeah, I assume thats peak. I wanted to overestimate the cost difference for affect. Even at 35 watts the cost savings per year for a single device isn’t much.
It more than what you think if its in a server farm.
You are forgotten. Its not just the cost of the server. Its the cost of the power supply to support it in case of power outtage and the size power feeds racks require. Each 1 watt in a server farm very quickly totals up to a huge number.
And a server farms have to work on the presume that peak will be required while in power outage.
Yes a server farm, obviously thats huge. No doubt. I think there are other reasons why it wont take off there as well, which I’ve explained in other threads in this discussion. This particular thread was about a single device in a small or home office. It makes no sense for MS to port to arm just for a small office server this thread was describing.
Bil,
I agree with you on the server farm concept. I do believe in 2 years we’ll be seeing a ton of rackmount servers that fit 40-80 cores in 1U that will run Linux or Windows and your favorite control panel and virtualization software of choice that will replace entire racks of web servers at hosting companies.
However, there have also been multiple attempts by vendors, specifically Cobalt/Sun/Oracle, Symantec, Cisco/Linksys, Apple, and small OEMs to build a small office server that is like an appliance that integrates all of the functionality that a business needs in one box. With the advent of cheaper SSDs and cloud functionality with cheaper bandwidth, this is entirely plausible and possible (and has been done before).
I understand the need for an all in one. I don’t understand why an all in one needs to be on ARM.
Because it’s one of the most prevalent for that form factor.
Still does not make sense. The form factor is not that important. Not important enough to convince Microsoft to port Windows Server 2008 and everything else over to ARM. Atom Boxes can get small. I don’t think small or home businesses would actually not buy such a useful device because it was a freaking inch too wide, high or deep for their tastes.
Bil,
Businesses buy things because of big names attached to them. With the right amount of marketing, you can get many small businesses to buy into a new business model.
This is just a device. However, Microsoft is getting their clock cleaned by Google, Apple, Amazon, and a large amount of Linux-based providers that offer cloud-based services to support small businesses. Cisco may finally be showing what they dropped all that cash for Linksys for at CES and over the calendar year instead of routers and wireless cards.
Microsoft’s business model of obscure licensing methods, rampant piracy of their products due to said licensing methods, the overall complexity of their products to set up and configure (It’s really hard for most people to admit, but Linux is getting really easy for most SMB tasks), especially compared to Cisco and Apple, and the requirement to have multiple servers to handle a simple small office is dying off.
It’s not that people can’t make everything work together. It’s that people aren’t going to do it if they don’t have to, and companies like Google give people that reason because they can.
If they don’t do something, Cisco, Google, Canonical, Apple, Red Hat, or a business based on their products will finish the job.
Microsoft needs to adapt to survive. Windows on ARM is a small piece of the overall puzzle in transforming their business model to allow for recurring revenue. It could be Windows on Atom, however, this is one small piece of what they need to do to finally break the silos and stop the bleeding.
YES ( sorry for the shouting, but nobody seems to understand what I’m saying)! I Agree with all of that, except I’m *completely* confused why ARM is even involved at all.
To me, its like arguing that water is wet, therefore everyone should eat more vegetables on Tuesdays. Yes water is wet, duh. Yes everyone should eat more vegetables. But why on Tuesdays? And what the hell does that have to do with the wetness of water?
Every time I say that it does not make sense, Someone ( well intentioned, I’m sure) Responds back arguing that water is indeed wet and vegetables are good for you and I’m crazy for thinking otherwise.
On some sense, its like I’m getting trolled by the universe in this thread.
Actually there should be more. This is an old 3-chip solution.
http://www.tomshardware.com/reviews/Intel-Atom-Efficient,1981.html
The desktop 945GC has a TDP of 22.2 W and the southbridge uses 3.3 W. Compared to the Atom 230 processor which has a TDP of 4 W and is supplied with a voltage of 1.088 V, this is a considerable difference.
22.2+3+4 = ~30W is only for motherboard
whole system: 40W idle – 44W full load
The energy savings will not even be worth the cost of planning the transition for the typical business. By the time ARM servers are cost effective x64 cpus will be even cheaper.
Sorry Thom but no. Not going to argue it except to point out that MS is a trend-follower, and the current trend is definitively *not* pointing in the direction of ARM servers, and *is* definitively pointing in the direction of ARM on portable devices.
Will you agree to eat your socks on this one?
It would be interesting if Microsoft could use a specially-tailored version of Windows Server software to build complex cloud of services / applications on a large cluster of low-priced servers based on cheap ARM processors. That could be VERY interesting…
A large score of ARM-based servers with no HDDs (only flash disks) would be cheap, low-energy AND it could also be powerful since I bet we could have those CPUs at relatively cheap price…
Windows on ARM? How should that bring an significant advantage to Microsoft? I don’t see a big issue porting the OS to another platform but all the applications? No way. It will take ages. The business world that is closely with Windows is not likely going to change quickly. Huge applications on Windows are mostly not written to be that portable. And even if they get ported to ARM, it will take a significant amount of time till they run as smooth as they run in x86. And on top of that keeping an x86 and an ARM version of an application takes time and resources. Look at the big players and how the struggle in getting applications to run on multiple architectures. IBM is such a good example. They offer for example their Lotus software line on many platforms and still they have not the same level of support for every platform. You can get Lotus Notes client for Windows, Linux and for Mac OS X. But some things are only offered on Windows (for example the Lotus Domino Administrator or Lotus Domino Designer). If it would be so easy as 1-2.3 to port an application to another platform (as many of you state here), then explain to me why IBM (without doubt one of the big players) does not port all their applications to every platform to the same extend as they do with Windows applications?
I personally just don’t see that huge advantage of having Windows on ARM? Probably the power savings are (for now) huger on ARM but most business dependent on Windows will not that easy change to something new. And those one that need to squeeze the last money out of hardware and software are anyway not the one running Windows. They mostly run either Linux or a BSD variant. Or is any one here thinking that Windows on ARM is going to push Google to switch to Windows servers? Probably not. Big data centers try anyway to decouple from Windows (if they can). Windows might have advantages (depending on what you need) but it is a OS with a not so low price tag. How could/should Windows on ARM change that picture?
You… didn’t read any of the comments in this thread. As such, I’m going to make fun of you.
“I personally just don’t see that huge advantage of having Windows on ARM?”
Is that a weird typo, a question to yourself, a call to telepaths to scan you, or an attempt at simulating up-speak? No matter which… nah.
Yes. It is a typo. Yeah, yeah. Make your fun
I have difficulty leaving schtick lying by the wayside
Windows NT has supported ARM, PPC, MIPS, and x86 for a long time (just look at the headers!), only Microsoft has only pushed out the x86 variant for over a decade. NT by design (though it’s HAL sub-system) support multiple architectures, and the VC++ compiler supports multiple architectures too. So don’t think this is much of an effort for Microsoft – it’s not. All it is is them saying that they are once again going to distribute Windows NT on another architecture.
You may be right about why – MS needs to start targeting tablets, netbooks, and other architectures that primarily use ARM instead of x86 processors.
So nothing new here, just move along.
Most Windows apps are now built using the .Net Framework, which could easily be ported to ARM, thus allowing all (err, most) .Net apps to run w/o any recompiling.
Yeah right. Business applications are build using the .Net framework. RDBMS? Messaging? Groupware? Web servers? etc… all of them are made mostly in .Net. Right? And they just run without recompile on ARM. Right?
Most .Net applications are compiled to full binaries and not .Net CLR Bytecode. So they would still need to be recompiled. .Net CLR Bytecode seems to be very seldom used and is more just a footnote in the .Net history than it is actual .Net usage.