IBM AltiVec

[arn]

Architosh posts (http://www.architosh.com/news/2002-09/2002a1-0903-64ppc-altvec.phtml) some speculation/interpretation about two posts regarding gcc (http://gcc.gnu.org/ml/gcc-patches/2002-08/msg01480.html).

The tentative conclusion is that IBM's PowerPC64's supports Altivec. Speculation of the new chip started after a description of an upcoming chip (http://www.macrumors.com/pages/2002/08/20020808110527.shtml) appeared on the conference schedule for Microprocessor Forum 2002 (http://www.mdronline.com/mpf/conf.html#day1_2). Since then, rumors of Apple-IBM talks (http://www.macrumors.com/pages/2002/08/20020809122153.shtml) over the new chip emerged. PenguinPPC (http://penguinppc.org/) initially reported that the vector unit on the IBM chip was Altivec compatible (http://www.macrumors.com/pages/2002/08/20020811060547.shtml), but that post has since been removed from PenguinPPC, which questions its original accuracy.

[bobky]

Woot!

[vniow]

Well, I'm not suprised. It's still good news. :)

[amnesiac1984]

will my brand new DDR powermac be upgradeable to 64bit IBM power4 based chips in a year or so? Otherwise I'll just shoot my own foot while eating my hat.

[avkills]

This is great news. I have no doubt that if IBM delivers, Apple will deliver and the Wintel world will be in tears.

8-way superscalar, 64bit, Alitvec and speeds up to 2Ghz. Bring it IBM, and let us put the final nail in the Wintel coffin.

-mark

[ffakr]

Originally posted by amnesiac1984
will my brand new DDR powermac be upgradeable to 64bit IBM power4 based chips in a year or so? Otherwise I'll just shoot my own foot while eating my hat.

I'd almost guarantee it is not compatable. IBM will not make a high speed 64 bit processor that runs on a 64bit MPX system bus.... Unless upgrade makers could do some daughter card hocus pocus (like getting a G3 to work on a 33MHz bus nubus machine). It'd castrate the chip though.

[vniow]

I'm not so sure that it wouldn't be just a pop in daughtercard. They redisigned the whole case, created a new motherboard with DDR, flipped it around so air can have a pass-through from front to back, a giant heatsink and fan. They didn't do this for no reason. The upgrades for all that are miniscule; a pair of 1.2 GHz G4 with 167 Mhz bus does not need that much modification. Ive got a feeling that this case is for something more.:)

[avkills]

I doubt the current Motherboards would support the chip to its full potential. For one, it looks like the chip uses HyperTransport for the FSB. But then again, Apple made the B&W motherboard so a G4 could just pop in.

My gut feeling is it would be a new motherboard. But then again, switching MB designs on the Xserve and the Current Towers so fast would be kinda strange I think.

Oh well, whatever happens, Apple had better use this chip whether it has Altivec or not. Motorola needs to be taught a lesson here.

-mark

[barkmonster]

I think 2Ghz is highly optimistic as a "starting at" speed, If they debut at 2Ghz maybe we'll see macs with 1.6, 1.8 and 2Ghz chips with a dual core 2Ghz chip as the Ultimate model so it's effectively a dual 2Ghz.

Apple are long overdue for a massive speed hike from either IBM or Motorola, if we're at 1.25Ghz right now and the jump from the PPC7410 to the PPC7450 gave us an increase in clockspeed of over 45%, a 2Ghz 64bit PowerPC chip with Altivec would be the perfect upgrade for the powermac line in February or March of 2003. 1.8Ghz wouldn't be too shabby either.

There's also the fact that these chips will be running most of the code in 32bit till people recompile applications that can take advantage of a 64bit chip, that gives us even more speed to look forward to in the future. It could be like the transition from G3 to G4, there were few applications that took advantage of altivec at first, after a year or so there were a lot of optimised applications using it. I can only imagine what a true DDR motherboard, Altivec AND a 64bit chip with fully optimised code could do for software synthesis and plug-in effects.

[Catfish_Man]

Originally posted by avkills
I doubt the current Motherboards would support the chip to its full potential. For one, it looks like the chip uses HyperTransport for the FSB. But then again, Apple made the B&W motherboard so a G4 could just pop in.

My gut feeling is it would be a new motherboard. But then again, switching MB designs on the Xserve and the Current Towers so fast would be kinda strange I think.

Oh well, whatever happens, Apple had better use this chip whether it has Altivec or not. Motorola needs to be taught a lesson here.

-mark

HyperTransport is not a system bus, it's for linking components on the motherboard together. IBM supports the RapidIO bus, which is similar to HyperTransport, but more in use in embedded processors (it's also better than HT according a very knowlegable person on arstechnica). Since this new chip is based on the Power4, I imagine it will use either a cut down version of the Power4's bus, or an on chip memory controller. Teaching Motorola a lesson would be stupid. Being stuck with IBM would be just as bad as being stuck with Motorola. Apple needs to have two suppliers competing. Using a chip without Altivec would be equally/more stupid, because 10.2 is MORE altivec optimized than 10.1 was, and not using Altivec would trash all the optimization work of the past 4 years (probably resulting in something that ran significantly slower than the current powermacs).

[daijones]

Read the rumors more closely and it sounds pretty bad for Apple. The rumors suggest that IBM is planning to release a Linux box under their own name. 1: an IBM brand linux box is going to be a big seller, taking sales off the power-hungry UNIX users Apple needs to be attracting; 2: if IBM want to sell this box, it would be in their interests NOT to sell the new chip to Apple, to reduce competition. I really hope Apple has maintained contact with IBM and has OS X compiled ready to put this chip into a box, because if IBM get much of a headstart with a 64-bit linux desktop the scientific market is lost to Apple before it even got going.

[noht*]

[shameless self-propaganda & view-count-fishing]
i told you so...
in this thread. (http://forums.macrumors.com/showthread.php?s=&threadid=10559) [/shameless etc.]

;)

noht

[synergy]

Originally posted by daijones
Read the rumors more closely and it sounds pretty bad for Apple. The rumors suggest that IBM is planning to release a Linux box under their own name. 1: an IBM brand linux box is going to be a big seller, taking sales off the power-hungry UNIX users Apple needs to be attracting; 2: if IBM want to sell this box, it would be in their interests NOT to sell the new chip to Apple, to reduce competition. I really hope Apple has maintained contact with IBM and has OS X compiled ready to put this chip into a box, because if IBM get much of a headstart with a 64-bit linux desktop the scientific market is lost to Apple before it even got going.

IBM high end boxes are just that, high end boxes. Not very cheap.
IBM sells CPUs to whoever will buy them as well. Apple is not in the high end server market either. The Xserve is not a high end server with max of dual processors. Some of those beasts come stacked with 10 or more processors.

[daijones]

The processor IBM are announcing in October is targetted at desktops and low-end servers: describes Apple perfectly. But it also describes a potential linux based range of products. I don't know any box manufacturers making PowerPC based boxes for linux, so I really hope the chip is going to Apple. It could be that IBM is thinking to use the chip in workstations, and also sell to Apple (and anyone who fancies becoming a linux box maker) to get the volumes up enough to make design/production more financially viable. But my point is, if alternatively IBM have decided to broaden their range of products to include low-end hardware, then this could be bad news for Apple, not good.

[MacCoaster]

For clueless people:

READ.

It says RS6000 down there. Hell:

Index: gcc/config/rs6000/linux64.h
===================================================================
RCS file: /cvs/gcc/gcc/gcc/config/rs6000/linux64.h,v
retrieving revision 1.25
diff -u -p -r1.25 linux64.h
--- gcc/config/rs6000/linux64.h 8 Aug 2002 17:55:37 -0000 1.25
+++ gcc/config/rs6000/linux64.h 24 Aug 2002 12:39:32 -0000

This is simply a diff file. --- means remove this line, +++ means add this line. Simply updated the stuff. Nothing big.

Doesn't mean it's for Macs or possibly for Macs.

As much as I want Apple to go IBM. Motorola is the only way they're going and we have to live with that.

[arn]

Originally posted by MacCoaster
For clueless people:

READ.

It says RS6000 down there. Hell:


So, what you are saying that the RS6000 has Altivec?

Still doesn't jive.

arn

[ddtlm]

MacCoaster:

I'm not sure how you conclude that IBM is not planning on offering a "Power-Lite" to Apple based on a two-line diff. Quite a leap of logic there.

[noht*]

about the rs/6000:
it can't 'have' or 'have not' altivec, as it's just a server architecture, not a processor; indeed, rs/6000 was rebranded the 'pSeries' more than two years ago. i guess it's still in the gcc config files for backward compatibility.

about the changes in gcc:
it's clear that this patch adds the altivec_target to the linux64 build of gcc. but why would this be done when there is no 64-bit processor that supports altivec? just for those users that want to compile osx/darwin/ppclinux apps on a rs/6000/pSeries...? kinda strange...

just some thoughts.

noht

[edit: "more than two years ago." -> it was fall 2000, actually...]

[AmigaMac]

I don't know any box manufacturers making PowerPC based boxes for linux, so I really hope the chip is going to Apple.

Actually there is one PPC/Linux vendor out there that is also an Apple reseller that builds its own boxes...

YellowDogLinux (http://www.yellowdoglinux.com/products/)

[avkills]

Originally posted by Catfish_Man


HyperTransport is not a system bus, it's for linking components on the motherboard together. IBM supports the RapidIO bus, which is similar to HyperTransport, but more in use in embedded processors (it's also better than HT according a very knowlegable person on arstechnica). Since this new chip is based on the Power4, I imagine it will use either a cut down version of the Power4's bus, or an on chip memory controller. Teaching Motorola a lesson would be stupid. Being stuck with IBM would be just as bad as being stuck with Motorola. Apple needs to have two suppliers competing. Using a chip without Altivec would be equally/more stupid, because 10.2 is MORE altivec optimized than 10.1 was, and not using Altivec would trash all the optimization work of the past 4 years (probably resulting in something that ran significantly slower than the current powermacs).

Ok, sorry about getting Hypertransport and RapidIO mixed up. All I know is it supports FSB transfers over 6GB/sec, which is damn fast.

I agree that using a chip w/o altivec for the Pro machines would be a mistake. But I also know that Motorola has been sucking wind lately. It is plain idiotic that the 750fx G3, which is in the iBook, supports speeds up to 1Ghz and a 200Mhz FSB bus that supports DDR and Apple will not use it to it's full potential. Altivec or not, I bet that thing would kick some serious booty running in that config.

Let's just hope that whoever supplies Apples next chip, it will be at or above 2Ghz, with a speedy FSB that supports DDR and some fast L2 & L3 cache.

-mark

[Rocketman]

Originally posted by avkills
This is great news. I have no doubt that if IBM delivers, Apple will deliver and the Wintel world will be in tears.

8-way superscalar, 64bit, Alitvec and speeds up to 2Ghz. Bring it IBM, and let us put the final nail in the Wintel coffin.

-mark

Technology is one thing but price is why so many crappy computers get sold to unsuspecting Wintel weenies.

If those crappy boxes are $200 less retail with a distributed distribution channel and 10% more margin for marketing, there is no hope against evil.

Rocketman

[DannyZR2]

What exactly does "8-way superscalar" mean????

[e-coli]

I know IBM as a company like the back of my hand. They've never made on single red cent on PC's. So a new low-end personal computer seems highly, and I emphasize highly, unlikely. It was just last year that IBM seriously considered selling off their personal computer division.

trust me. ;) No threat to Apple.

[avkills]

Originally posted by DannyZR2
What exactly does "8-way superscalar" mean????

I think it affects how many operations can be performed in a single clock cycle. I could be wrong though. The current G4 is 4-way superscalar.

-mark

[dongmin]

Originally posted by e-coli
I know IBM as a company like the back of my hand. They've never made on single red cent on PC's. So a new low-end personal computer seems highly, and I emphasize highly, unlikely. It was just last year that IBM seriously considered selling off their personal computer division.

trust me. ;) No threat to Apple.

I don't think anyone is suggesting that IBM is getting into the "low-end personal computer" market, but rather the high-end desktop market and low-end [I]server[/B] market. One possible strategy might be the consumer Macs being outfitted with Motorola G4s and the pro desktops being outfitted with the Power4-variant.

[Tiauguinho]

Originally posted by DannyZR2
What exactly does "8-way superscalar" mean????

A 8 way Superscalar Processor Chip means that it can execute 8 instructions per clock cycle. The current G4 is a 4 way Superscalar (4 instructions per cock cycle). But you have to have this on mind. If you have instructions that depends on the result of the previous instruction, the Superscalar technology will not bring you any benefit, because it processes/execute 8 instructions that are independent form each other. You cant have the chicken and the egg at the same time...

[kenohki]

Originally posted by daijones
Read the rumors more closely and it sounds pretty bad for Apple. The rumors suggest that IBM is planning to release a Linux box under their own name. 1: an IBM brand linux box is going to be a big seller, taking sales off the power-hungry UNIX users Apple needs to be attracting; 2: if IBM want to sell this box, it would be in their interests NOT to sell the new chip to Apple, to reduce competition. I really hope Apple has maintained contact with IBM and has OS X compiled ready to put this chip into a box, because if IBM get much of a headstart with a 64-bit linux desktop the scientific market is lost to Apple before it even got going.

Why would they want to cannabalize their sales of AIX ($$$) with sales of Linux (free)? Yes, they've done lots of Linux on a Mainframe (zSeries) but does this make sense in their traditional Unix space (pSeries / RS/6000). I would think this is more just an update to replace the 604.

Also, a lot of the scientific market is not gaga over Linux. I used to work in the chemistry department at Argonne National Laboratory and they'd rather use Macs provided they can get equivilant power out of them. Many of those guys have to do a lot of their IT work themselves (including programming) and they don't want to have to screw with KDE and tweak X just to get a usable environment. Anyway, just some thoughts.

[synp]

IBM has had a Linux strategy for the last three years - provide Linux on all hardware platforms so that applications can smoothly scale from the low-end xSeries to the high-end zSeries and everything in between.

This was part of their "eBusiness" strategy (they actually tried to trademark the "e"). To show what a great e-business company they were, they renamed all the server platforms. That's why the Intel servers became "eServer xSeries", the RS/6000 became "eServer pSeries", the AS/400 became "eServer iSeries" and the mainframes became "eServer zSeries". They want to support Linux on all these platforms in addition to the legacy OS with names like Windows, AIX, OS/400 and MVS (renamed z/OS). That's why it makes perfect sense for IBM to support Linux on a pSeries.

As for the new PPC, it could be destined for a low-end pSeries which should run either Linux or AIX just as all the other pSeries servers do. They will never compete with the iMacs and eMacs, because they don't make money on PCs. There may be some competition between low-end pSeries and PMs, but IBM has a long history of selling to competitors. IBM's biggest focus is Sun with whom they compete for the mid-range and high-end Unix servers, where margins are high. Apple is to them more of a customer than a competitor, since it operates in a low-margin market.

[technocoy]

I hate to be vague, I have sources through IBM.... Motorola has been dumped.:cool:
don't wanna get my butt into hot water... sorry

[Faeylyn]

Originally posted by daijones
But my point is, if alternatively IBM have decided to broaden their range of products to include low-end hardware, then this could be bad news for Apple, not good.

Linux is no threat to Apple. Period. There are currently many many many many different Linux boxes out there. They range in price from very low (Walmart's new $199 box comes to mind) to very very expensive. They range in power from very low, to many many times as powerful as the best Apple has to offer. These are all available TODAY. These have all been available for quite a while. So suppose IBM comes out with a Linux box. So what? How would that be different than any other Linux box out there? Why would people who don't want to use Linux suddenly shout for joy because there's a new box that happens to have a Power4-Lite inside?

IBM's main competition for a Linux box would be other Linux vendors, not Apple. IBM would have to convince the people looking for Linux boxes that IBM's solution offered a better price/performance value. Just making it faster won't cut it. IBM would have to make the $ per flop significantly less in order to get people to buy the thing. This means keeping margins low.

As for IBM not selling the chips to Apple, wrong, wrong, and dead-wrong. Apple probably sells more computers in a single week than an IBM/Linux PC would sell in a whole year. There is no way IBM would pass on that kind of cash. If Apple's sales take off, it's good news for IBM.

[e-coli]

Originally posted by technocoy
I hate to be vague, I have sources through IBM.... Motorola has been dumped.:cool:
don't wanna get my butt into hot water... sorry

the campus in raleigh is IBM Global Services, and has almost nothing to do with OEM, or chipmaking technologies. I would venture to bet that if your "source" is from the RTP campus, his / her info is rooted in speculation.

[chicagdan]

All of these IBM G5 stories are great ... but there's no way in hell I'm buying a new Apple in 2002 if I can buy a G5 equipped system in 2003. In a deflationary industry like desktop computers, I'd be nuts to buy a 17" iMac today if I can get the same configuration with a 2 GhZ G5 one year from now ... I can muddle through another year with my current computer.

That's the problem of having computers priced so much higher than the rest of the industry -- you force consumers to wait for a better deal. Plus Apple systems are so well built that my souped up Beige runs Jaguar just fine. The only reason to buy a new system right now is for the aesthetic beauty of the iMac.

[oldMac]

The *only* reasons I can think of for IBM to make an Altivec-enabled chip are the following:

1) To sell to Apple

2) To attempt to optimize certain Linux applications in a way that differentiates them from the Intel-driven Linux world so that there's a reason for people to buy expensive Linux hardware from IBM.

Neither reason seems like enough of a reason for IBM to bother on its own. I'm guessing that both reasons are accurate.

[kenohki]

Originally posted by synp
IBM has had a Linux strategy for the last three years - provide Linux on all hardware platforms so that applications can smoothly scale from the low-end xSeries to the high-end zSeries and everything in between.

Yeah, and theoretically this makes a lot of sense. Especially if the OS we were talking about was very scalable. I guess I just question Linux's scalability. I mean it's pretty limited to small system images and clustering. On the zSeries they've been basically partitioning the machine into virtual machines each running Linux. I'm sure that makes for some very easy machine administration versus having tons of small boxes to have to maintain. But it's nowhere near as elegant or powerful as z/OS, MVS, OS/390...whatever they're calling it these days.

As for the pSeries, I can see Linux running on a low to mid-range PPC box, but I can't see it scaling up to something like the p690. And when the FastPath stuff gets integrated into POWER5 and POWER6, is some of that intellectual property going to go back into Linux? Guess it depends on how it's implemented. Just a rhetorical question. *shrug*

[technocoy]

I LOVE APPLE!

Later

[3G4N]

Originally posted by technocoy
I hate to be vague, I have sources through IBM.... Motorola has been dumped.:cool:
don't wanna get my butt into hot water... sorry

while trying to stay out of that same
vague james brown hot tub,
I have sources at Motorola
saying the exact same thing.

[eirik]

How long did we wait for our favorite apps to implement Velocity Engine? Point is, while gcc and perhaps other compilers are now making implementation easier, it still reqires quite a bit of effort to lever AltiVec.

IBM says that junior is targetted for workstation and low-end server markets. Could someone in the know please let us know what major multimedia packages work on linux. I'm not actually requesting a comprehensive listing, just a qualitative feel for what is out there.

AltiVec is great for multimedia apps and for scientific calcuations other than perhaps double-precision floats. Perhaps, IBM's SIMD will not only be compatible with Motorola's implementation of VMX but it also adds support for SIMD on double-precision data. Wouldn't that be something!

If not obvious already here, who would IBM seek to sell IBM branded Linux boxes to? These boxes would most likely be considerably more expensive than other Linux boxes on the market. Clearly, IBM would seek a means to differentiate its boxes from those of the others. Their VMX implementation could perhaps. But, who would buy them?

As far as I know, Photoshop doesn't run on Linux. The other major multimedia (high-end) apps run on Solaris and SGI's OS (sorry brain fart, can't remember, dah!). Well, porting these apps to Linux wouldn't be very difficult. But, enabling them to employ junior's SIMD would likely be challenging to say the least, unless the new compilers are making it a hell of a lot easier.

Even so, the consumers of this market are pretty damn high end, do they really care that much if its Linux instead of AIX? I don't know how IBM license pricing compares but why invest in this porting if it might also enable other Linux box makers to enjoy the fruits of the efforts minus AltiVec.

I don't presume to have all of the answers. However, this concept just seems rather fuzzy to me.

Eirik

[nixd2001]

Originally posted by technocoy
I hate to be vague, I have sources through IBM.... Motorola has been dumped.:cool:
don't wanna get my butt into hot water... sorry

Wow. Do you know if this extends across the range or just for top-end processors?

And there's the thread elsewhere about Apple shares being bought by some corporate.......

[Cappy]

Originally posted by kenohki


Why would they want to cannabalize their sales of AIX ($$$) with sales of Linux (free)? Yes, they've done lots of Linux on a Mainframe (zSeries) but does this make sense in their traditional Unix space (pSeries / RS/6000). I would think this is more just an update to replace the 604.

They've done things like this before or do you not remember IBM selling systems with OS/2 and others with WinNT? They went with what the market demanded because initial sales mean more to them. They're doing the same thing again only with Linux.

Personally I'd love to see IBM sell a highend server solution that included Mac OS. Probably won't happen in my lifetime but would be a boost to Apple in the business world nonetheless.

[jaykk]

IBM could use their own chip if they want to. Think about all those thinkpads with Win installed? If ibm can supply thinkpad with Linux, it is the end of Microsoft .. I heard that Sun already replaced winnt laptops with Apple Powerbooks. and HP chose Corel WordPerfect over M$ Office. If IBM follows the suite with Linux installed thinkpads, it is indeed a great news for anti-m$ campaign.

[Faeylyn]

Originally posted by Cappy

Personally I'd love to see IBM sell a highend server solution that included Mac OS. Probably won't happen in my lifetime but would be a boost to Apple in the business world nonetheless.

That's actually a really good idea. Apple could even design the thing. And since it's a high-end server, it wouldn't be competing with anything Apple has or is likely to have for the next several years. It could even include the Power4 for that matter and later the Power5 and Power6. And since high-end servers have high-end profit margins, IBM could give Apple a healthy royalty on each hardware unit sold.

[Faeylyn]

Originally posted by jaykk
If ibm can supply thinkpad with Linux, it is the end of Microsoft .. I heard that Sun already replaced winnt laptops with Apple Powerbooks. and HP chose Corel WordPerfect over M$ Office. If IBM follows the suite with Linux installed thinkpads, it is indeed a great news for anti-m$ campaign.

1. Not knowing the specifics yet, it's doubtful that the first generation of the Power4-Lite would be useful in a laptop.

2. Linux is not a threat to Microsoft in the consumer arena. You can run Linux on laptops TODAY. Why don't most people do this? Are people who don't want to run Linux on an x86 laptop likely to want to run it on a Power4-Lite laptop? What would they gain?

3. There are lots of laptop makers besides IBM. Most people prefer the other brands.

[ktlx]

Originally posted by Faeylyn

That's actually a really good idea. Apple could even design the thing. And since it's a high-end server, it wouldn't be competing with anything Apple has or is likely to have for the next several years. It could even include the Power4 for that matter and later the Power5 and Power6. And since high-end servers have high-end profit margins, IBM could give Apple a healthy royalty on each hardware unit sold.

Why would IBM, who has experience building high end servers, contract to Apple, who has almost no experience building servers of any kind, to build a high end server? That makes no sense. Mac OS X is not even very competitive against low end server offerings in terms of features you expect out of an enterprise-class server operating system. It does not have a journaling file system (geez, Apple even recommends to use HFS+ over ufs), it does not support enterprise backup solutions and its disk subsystem options are still pretty limited.

[skunk]

If Motorola are out of the door, presumably that means IBM for January? And what happens to the iMac/eMac/TiBook? How will they be upgraded?

[Newtonboy]

Originally posted by edvniow
I'm not so sure that it wouldn't be just a pop in daughtercard. They redisigned the whole case, created a new motherboard with DDR, flipped it around so air can have a pass-through from front to back, a giant heatsink and fan. They didn't do this for no reason. The upgrades for all that are miniscule; a pair of 1.2 GHz G4 with 167 Mhz bus does not need that much modification. Ive got a feeling that this case is for something more.:)

The problem I have with this idea is Back when all the upgrade companies started going belly up, I heard it was because Apple considers upgrade suppliers as competitors. So why would Apple help out the competition by making a case for their use?
Then again I could be wrong, and Apple could have learned that people their macs and don't give them up for a whole new machine too easily. (Or at least I don't) They upgrade them, and customize them, and make'um their own.

[eirik]

Originally posted by ktlx


Why would IBM, who has experience building high end servers, contract to Apple, who has almost no experience building servers of any kind, to build a high end server? That makes no sense. Mac OS X is not even very competitive against low end server offerings in terms of features you expect out of an enterprise-class server operating system. It does not have a journaling file system (geez, Apple even recommends to use HFS+ over ufs), it does not support enterprise backup solutions and its disk subsystem options are still pretty limited.

I'm a bit skeptical about the utility/importance to server administrators of a journalling file system. Nonetheless, ktlx raises a good point to me. I'd like to see a thread digging into what features OS X lacks for various server markets/needs. There we could discuss, for examlpe, the utility/need for a journalling file system in a server world.

I'd start it myself but I don't have that priviledge in Macrumors.

Eirik

[ktlx]

Originally posted by eirik


I'm a bit skeptical about the utility/importance to server administrators of a journalling file system. Nonetheless, ktlx raises a good point to me. I'd like to see a thread digging into what features OS X lacks for various server markets/needs. There we could discuss, for examlpe, the utility/need for a journalling file system in a server world.

I'd start it myself but I don't have that priviledge in Macrumors.

Eirik

The first time you have a hard crash of a 500MB file system on a live system you will see the importance of a journaling file system. :D Without a journaling file system, the file system will either come up possibly corrupt or you will need to do a fsck on it. If it is corrupt and you try to write, the file system is toast and it is time to get out your backup tapes (RAID does not help if the file system is corrupt). If you run fsck, your server is unavailable until it is complete. Large file systems can take a long time to run fsck on. I have seen file servers without journaling file systems take hours to come up waiting on their fscks to complete.

With a journaling file system, this does not happen (except in rare cases that would have trashed a non-journaling file system anyway). The file system is either kept in a non-corrupt state or a recoverable state by backing out incomplete updates. For an enterprise-class database server or large file server, this is critical. Web and print servers can get away without a journaling file system just fine. Most e-mail systems could probably get away without it, although if it is a large one, having the mail spool and storage file systems journaled is nice.

These features are so critical to enterprise class servers that Veritas built a very successfuly business solely around providing journaling file systems and software RAID for UNIX systems. When Linux first started going after the server market, one of the knocks against it was that it had no (and no one provided) a journaling file system for it. Since then several companies stepped in to fill in the gaps. Linux still has some limitations, but it is much farther along than Mac OS X.

Please don't get me wrong, the Xserve is a nice server and Mac OS X is a nice operating system. It is a wonderful system for a small operation or for most Web, file and print services. But to me, a high end server is an enterprise-class server that is going to run a database (i.e. Oracle or DB/2), application server (i.e. BEA, PeopleSoft or SAP) or large e-mail or file server. Apple has no experience here and Mac OS X is woefully inadequate compared to Solaris, HP-UX, AIX and Linux. And Linux is only on that list, in my opinion, thanks to the huge investment by IBM.

It will very difficult for Apple to get much traction in the large server market as Sun, Hewlett-Packard and IBM own 90% of the UNIX server market. All of them have far more experience than Apple at building large servers. My guess is that as Linux matures and Intel improves their server CPU designs, Dell will capture the remaining market share leaving only crumbs for Apple or any other company.

[Cappy]

Originally posted by ktlx


The first time you have a hard crash of a 500MB file system on a live system you will see the importance of a journaling file system. :D Without a journaling file system, the file system will either come up possibly corrupt or you will need to do a fsck on it. If it is corrupt and you try to write, the file system is toast and it is time to get out your backup tapes (RAID does not help if the file system is corrupt). If you run fsck, your server is unavailable until it is complete. Large file systems can take a long time to run fsck on. I have seen file servers without journaling file systems take hours to come up waiting on their fscks to complete.

With a journaling file system, this does not happen (except in rare cases that would have trashed a non-journaling file system anyway). The file system is either kept in a non-corrupt state or a recoverable state by backing out incomplete updates. For an enterprise-class database server or large file server, this is critical. Web and print servers can get away without a journaling file system just fine. Most e-mail systems could probably get away without it, although if it is a large one, having the mail spool and storage file systems journaled is nice.

These features are so critical to enterprise class servers that Veritas built a very successfuly business solely around providing journaling file systems and software RAID for UNIX systems. When Linux first started going after the server market, one of the knocks against it was that it had no (and no one provided) a journaling file system for it. Since then several companies stepped in to fill in the gaps. Linux still has some limitations, but it is much farther along than Mac OS X.

Please don't get me wrong, the Xserve is a nice server and Mac OS X is a nice operating system. It is a wonderful system for a small operation or for most Web, file and print services. But to me, a high end server is an enterprise-class server that is going to run a database (i.e. Oracle or DB/2), application server (i.e. BEA, PeopleSoft or SAP) or large e-mail or file server. Apple has no experience here and Mac OS X is woefully inadequate compared to Solaris, HP-UX, AIX and Linux. And Linux is only on that list, in my opinion, thanks to the huge investment by IBM.

It will very difficult for Apple to get much traction in the large server market as Sun, Hewlett-Packard and IBM own 90% of the UNIX server market. All of them have far more experience than Apple at building large servers. My guess is that as Linux matures and Intel improves their server CPU designs, Dell will capture the remaining market share leaving only crumbs for Apple or any other company.

You may be right but some in the business do see some opportunites for a fully featured Mac OS X Server to make some inroads. I would agree that Apple has no business being in the hardware market for highend servers. Software is another matter and lets face facts. Nothing says that this would happen tomorrow. It's obvious that Apple isn't ready yet.

As for not having experience there is always a first and, sure, the market is pretty full but with the right features and price, IBM could sell this thing. They still sell OS/2 last I checked. I don't see anyone else capable of pulling this off. Not Dell, Sun, HP, or even Apple....only IBM.

[ktlx]

Originally posted by Cappy
You may be right but some in the business do see some opportunites for a fully featured Mac OS X Server to make some inroads. I would agree that Apple has no business being in the hardware market for highend servers. Software is another matter and lets face facts. Nothing says that this would happen tomorrow. It's obvious that Apple isn't ready yet.

As for not having experience there is always a first and, sure, the market is pretty full but with the right features and price, IBM could sell this thing. They still sell OS/2 last I checked. I don't see anyone else capable of pulling this off. Not Dell, Sun, HP, or even Apple....only IBM.

You are talking could and I am talking should. Sure it is possible for IBM to contract out to Apple to build software, hardware or whatever for the server market and try to sell it. But the question is why? What reason would IBM have to believe that Apple could develop any product targetted toward the enterprise server market that would be more attractive than something IBM developed itself?

The problem with the catch up argument, in my opinion, is one of R&D dollars. The server portion of Apple's business, let alone any potential high end server business, would be so small in comparison to the existing players, they will simply be out spent. Before we even talk about things like journaling file systems we need to find out if Mac OS X will even work with more than two processors. It should, but others had problems taking SMP beyond two processors before. Until there is a quad or eight processor Mac OS X server out there running just fine, we will never know for sure.

[Faeylyn]

Originally posted by ktlx

Why would IBM, who has experience building high end servers, contract to Apple, who has almost no experience building servers of any kind, to build a high end server? That makes no sense. Mac OS X is not even very competitive against low end server offerings in terms of features you expect out of an enterprise-class server operating system. It does not have a journaling file system (geez, Apple even recommends to use HFS+ over ufs), it does not support enterprise backup solutions and its disk subsystem options are still pretty limited.

Because it wouldn't just be a high-end server, it would be a high-end Mac server. IBM has zero experience with Macs and there's no way they'd be able to do that without Apple's help.

As for the rest, you're thinking in terms of what traditional servers are used for. Again, this isn't a high-end server, it's a high-end Mac server. Not the same thing, not the same uses, not the same customers.

[Faeylyn]

Originally posted by ktlx

But to me, a high end server is an enterprise-class server that is going to run a database (i.e. Oracle or DB/2), application server (i.e. BEA, PeopleSoft or SAP) or large e-mail or file server. Apple has no experience here and Mac OS X is woefully inadequate compared to Solaris, HP-UX, AIX and Linux. And Linux is only on that list, in my opinion, thanks to the huge investment by IBM.

It will very difficult for Apple to get much traction in the large server market as Sun, Hewlett-Packard and IBM own 90% of the UNIX server market. All of them have far more experience than Apple at building large servers. My guess is that as Linux matures and Intel improves their server CPU designs, Dell will capture the remaining market share leaving only crumbs for Apple or any other company.

Again, I'm not talking about a high-end server, but a high-end Mac server. Macs don't compete in the areas you mention above. There's no reason for Apple to enter that market because there's already enough competition there that Apple would find it difficult to impossible to differentiate their offerings.

Instead, you try to marry IBM's existing knowledge about how to build a decent high-end server with Apple's OSX for a higher-end Mac server than Apple can produce. Apple would have to be involved in the design because it's a Mac server, not just another server.

Initially, this would be for people who already want Mac servers but find the XServe more than lacking. How big is that market? Or, a better question, how big is that market likely to be two years from now?

[ktlx]

Originally posted by Faeylyn


Again, I'm not talking about a high-end server, but a high-end Mac server. Macs don't compete in the areas you mention above. There's no reason for Apple to enter that market because there's already enough competition there that Apple would find it difficult to impossible to differentiate their offerings.

Instead, you try to marry IBM's existing knowledge about how to build a decent high-end server with Apple's OSX for a higher-end Mac server than Apple can produce. Apple would have to be involved in the design because it's a Mac server, not just another server.

Initially, this would be for people who already want Mac servers but find the XServe more than lacking. How big is that market? Or, a better question, how big is that market likely to be two years from now?

Okay, but you still have not answered the question why would IBM want to be involved beyond building the CPUs for this thing. So we have this high end server running Mac OS X server, not targeted toward the cash-cow enterprise market, but some other market such as render farms, compute servers, etc.

If I am IBM what reason is there for me to be involved beyond wanting Apple to use my Power4-lite (or other PPC) processors? What reason is there to believe the market of buyers who must have Mac OS X, as opposed to Linux, is worth messing with for IBM? Why would many people even want a high-end server based on a 32-bit operating system that is not Microsoft Windows?

I am sure there is a market for a server with more muscle than the Xserve running Mac OS X server. My contention is that the size of the market looking for an Apple designed and IBM marketed high end server, is too small for IBM to even mess with. That is assuming Apple could even design and get working a four, eight or more CPU server using Mac OS X. Xserve is more an effort in industrial design than anything else (they already had dual processor systems).

[Faeylyn]

Originally posted by ktlx

Okay, but you still have not answered the question why would IBM want to be involved beyond building the CPUs for this thing. So we have this high end server running Mac OS X server, not targeted toward the cash-cow enterprise market, but some other market such as render farms, compute servers, etc.

If I am IBM what reason is there for me to be involved beyond wanting Apple to use my Power4-lite (or other PPC) processors? What reason is there to believe the market of buyers who must have Mac OS X, as opposed to Linux, is worth messing with for IBM? Why would many people even want a high-end server based on a 32-bit operating system that is not Microsoft Windows?

I am sure there is a market for a server with more muscle than the Xserve running Mac OS X server. My contention is that the size of the market looking for an Apple designed and IBM marketed high end server, is too small for IBM to even mess with. That is assuming Apple could even design and get working a four, eight or more CPU server using Mac OS X. Xserve is more an effort in industrial design than anything else (they already had dual processor systems).

Why would IBM want to be involved? The same reason they want to be involved in anything they do. Money. They could gouge high-end customers in the same way they currently do with their other mid to high-end servers. And then make several times that again on the support fees. I don't have IBM's sales figures, but they certainly don't need to sell a huge number of units of higher-end servers.

As for it being a 32-bit OS, my bet is that will change sometime in 2003.

Thinking longer term, the more Apple boxes (with IBM chips) that sell, whether they be desktop or server, or IBM-built server, the better for IBM. How much money does IBM get when someone buys a Dell or Gateway or Compaq or HP or Sun? Unless they're using a couple IBM components, the answer is zero to none. If supporting Apple on the high-end helps drive the low-end, that's good for IBM twice (gouge the high-end customers, make a few bucks on the chips used for the low-end customers). If putting IBM's name on a Mac OSX box helps gets Macs in the doors of more businesses, schools, etc., that's nothing but good news for IBM. Even if it just adds a little respectability to Mac OSX, it still is a win for IBM.

But mostly it's because I want them to. :)

[Faeylyn]

This is similar, in a way, to Intel's advertising. Intel wants to encourage people to use systems with Intel processors because they don't make money on systems with AMD, PowerPC, or anything non-Intel. I've always wondered why IBM and Motorola didn't do more to help push their processors to the public.

IBM should want to do whatever it can to encourage people to buy Macs. When people buy Macs, IBM makes money. Just selling the CPU to Apple with a, "thank you, come again" isn't very good business sense.

So if Apple needs a boost to get into corporations, IBM should help. If putting Mac OSX on an IBM branded server would help Apple get their foot in the door in other ways, it would be a win for both companies.

[Cappy]

Originally posted by ktlx


You are talking could and I am talking should. Sure it is possible for IBM to contract out to Apple to build software, hardware or whatever for the server market and try to sell it. But the question is why? What reason would IBM have to believe that Apple could develop any product targetted toward the enterprise server market that would be more attractive than something IBM developed itself?

Manageability, reliability, security. Again I'm not saying this could happen tomorrow but Apple has a better base to build off of than even Linux. Let us not forget costs which is where Linux makes inroads. Mac OS X Server would admittedly cost more than Linux but should still come in cheaper than a MS solution while still offerring better features over Linux.

Originally posted by ktlx
The problem with the catch up argument, in my opinion, is one of R&D dollars. The server portion of Apple's business, let alone any potential high end server business, would be so small in comparison to the existing players, they will simply be out spent. Before we even talk about things like journaling file systems we need to find out if Mac OS X will even work with more than two processors. It should, but others had problems taking SMP beyond two processors before. Until there is a quad or eight processor Mac OS X server out there running just fine, we will never know for sure.

I agree with this one. SMP beyond two processors is sorely needed. Journaling too would be a great benefit but I don't see that as being a big decision maker as the SMP. It would certainly help and by the time some of this came to fruition, I wouldn't be surprised to see a journaling file system.

At any rate you are preaching the idea of it likely not happening to the choir. We all know that Apple has no experience in this arena. My point that brought it up was that I felt that if they focused on bringing a more solid product to address that market, IBM would be the one that could put Apple on the map in a business environment. No one else can do this in my opinion.