Slashdot Mirror
Alpha Release Of Red Hat's Itanium Distro
nicktook writes: "ZDNet has this story that RedHat has released a full alpha distribution for the 64-bit Itanium processor. Can Itanium hardware even be bought yet?" Not by jes folks, that's for sure. cnoe also sent a link to the official announcement from Red Hat as well. Coupled with SGI's release of Pro64 compilers for Itanium, it seems like Intel's Next Big Chip keeps whispering "penguins." Stunning news from MS on this front is long overdue.
On the other hand, I'm seeing a lot of posts saying that this is going to 'make' Linux. I'm sorry kids. I've said it before, and I'll say it again, nothing is going to 'make' Linux all by itself. Awareness is the only thing that will 'make' Linux.
How many people were hit by the I.Love.you virus? Probably close to the same number of people that have learned to accept such 'inconveniences' in day to day life. Sure, we wouldn't put up with it, but we don't have to. We know other things. Anyway, I'm getting offtopic...
What I'm trying to say, is that the average, every-day, Joe Blow Consumer doesn't care about proc optimizations and compatability. If they hear that Microsoft doesn't have one running Itanium yet, they'll wait to buy Itanium chips. When MS is ready, so will they be. The chances they'll run across an Itanium with RHat on it at their local Circuit City, or CompUSA, or Best Buy is slim to none. And even if they do, they're not going to bite. They'll wait for MS to push them towards Itanium. And then they'll go out and plunk out the cost of yet another OS upgrade, and won't pay it a second notice.
This is our sad reality... Accept it. No single feature, optimization, or compatibility will 'make' Linux. We're not going to sweep Microsoft under the rug with this. We may gain a few users, and that's fine with me. Every little bit counts. Once people become aware that there is an operating system out there that doesn't crash every day, or run slower as they day progresses, they won't want to go back. At that point it will become worth the effort to learn a new OS.
Take it how you want.
First it states it is an Alpha release and later an IA-64 release; which one is it?
:-) = I am happy
:^) = I am happy with my big nose
C:\> = I am happy with my OS
We are simply less organized, and less motivated, because it's what most of us do in our SPARE time, instead of full time. That is not a detriment though. I submit that because most of our work is done in our off hours, we take more pride in it. Which is why our code doesn't have 63,000+ bugs.
Linux, on the other hand, gets ported quickly, and almost by habit to new platforms.
Again, 100% correct. Portability is a huge factor. MS doesn't have to worry about portability as much, because whatever they port it to is going to be their own OS, which (one should assume) has backword/forward compatability built in. The portability of OS however, means that when one company makes a great leap, as RHAT did in this case, then we all jump right along with it. What works for RHAT works for me. I may have to mod the code a little, but then it's re-released for whatever I ported it to.
I'm not arguing with anything you said, because I agree almost whole-heartedly. I still just think that it's 'awareness' that is our 'one big thing'. That lack thereof is all that's hurting us. A lot of people don't want to learn new things, but some don't mind. Some of them are in positions of power.
Example: Bill the Bossman is a manager at a data entry facility. Bill notices Linux at his brother's house. Bill's brother runs uptime, and shows Bill that his Linux machine has been used heavily, but not rebooted for the past 14 days.
Bill thinks, "Wow... This could certainly change things at MegaCorp... no more rebooting and losing work in the middle of the day. That I LOVE YOU virus cost me three weeks worth of hours!" Bill heads back to MegaCorp and checks with his IT dept. The IT dept thinks switching is a great idea. 250 machines are switched over, and employees trained.
Because most of their work is data entry, only StarOffice is focussed on, and people are now running Linux. These same people, while still clueless as to what's going on, notice that they're not crashing. They no longer see the BSOD. Their work is safer, and they are more productive. They then check into Linux, etc...
In another comment, SweenyTod notes that people want to use the same applications between work and home, and he's also right. So if they're using Linux at work...
Anyway, sorry for the length of this post. It wasn't necessary, but it's a slow night here, and I'm bored as hell. Thanks for listening...
> I just wonder how sucessfull ia64 is going to be. ... It will proberbly be huge, just to spite me, but let me just say that I wont be a early adopter!!
I doubt that there will be much consumer rush for it. Even if the Win2K port runs well on it, Win2K holds a very tiny share of the world's desktops right now. Nor is the Itanium likely to be cheap. I'm sure Intel will push it for servers, but even there it will be interesting to see how its price:performance ratio stacks up against the latest generation of x86 systems.
I suspect 32-bit x86 will reign the desktop for quite a few more years, except among scientists and engineers. And as you say, it will be interesting to see how IA-64 competes among those CPU-power-users. (At least with Linux they can pick an architecture based on its price:performance ratio, rather than being stuck with the one that their store-bought operating system happens to support.)
> I tried linux on a Digital Alpha and this _cured_ me of most of my desire to try alternative platforms.
I presume the things that didn't work for you were the various utilities and applications, rather than Linux per se? Do lots of open source apps break even if you recompile them? I'd be interested in hearing a bit more about your experience.
--
Sheesh, evil *and* a jerk. -- Jade
IA64 is a big deal. Intel gets better yields than any other chip company. They will ship massive quantities of IA-64's and price Alpha's and SPARC's out of the market.
IA-64's can beat out competitors than same way previous Intel/MS combinations sent Apple to 15% market share. Gaggles of vendors (including Compaq) hocking IA-64 based machines will put Sun and Alpha out to pasture. If you look at the history of the computer industry you will see that technological supieriority does not equal market dominance.
Conformity is the jailer of freedom and enemy of growth. -JFK
linuxdev.net about the ottawa linux symposium mentioning linux on merced
a comment on the linux kernel list regarding SCO Unix running on IA-64 as early as 1997
and of course you should check out the Linux on IA-64 project homepage.
greetings, eMBee.
--
Gnu is Not Unix / Linux Is Not UniX
--
--
We have fought the AC's, and they have won.
> Stunning news from MS on this front is long overdue.
I dunno. I was stunned by several news stories about Micorsoft over the last couple of weeks. Promiscuous love letters, the Cookie Monster, and now sordid revelations about Clippy's secret life.
What Micorsoft needs right now is longer gaps between stunning revelations.
--
Sheesh, evil *and* a jerk. -- Jade
IA-64 is a LIW chip, not RISC. Unfartunately I don't know if this will allow it to be faster (probably largely dependent upon the instruction types being used in a given app) ie - might be faster for graphics processing but slower or equally fast as an Ultra Sparc for something like networking.
----
Slán leat agus go n'eirí an bóthar leat
I'm curious about this. I've been a long time user of Sparc Linux, and I've found virtually no differences. The obvious ones I have noticed are the sexy console (which x86 can finally now do with the framebuffer support), and the fact that x86 binary-only apps (e.g., Acrobat Reader, Oracle etc.) obviously won't work. I can happily run the SunOS version of Acrobat Reader, though, and I'm not too bothered about other closed source apps. The only problems I've had with open source programs have been those that include x86 assembler with no C alternatives. What problems did you have with non-x86 Linux?
"The invisible and the non-existent look very much alike." -- Delos B. McKown
Having worked with said SDK, the compilers are certainly available, but the current state of Windows on IA64 is a joke. Most of the management tools don't work, just getting the latest beta installed took 4 tries since the installer keeps crashing and the promised 32 bit compatibility is, for us at least, far from good. What's particularly disappointing is that Windows doesn't seem to have advanced much since the last beta we got months ago. They don't even have a self-hosting compiler yet which makes development a severe PITA. I don't know how Linux stacks up against this yet, but what's out there for Windows right now doesn't really qualify as what I would call 'usable.' At least the latest beta lets you make shortcuts (trying to do so on the last one would crash explorer). So I guess it all depends on what you would define as a 'release' (although as I said, I have no idea what the state of the Linux Itanium world is, it's very possibel I wouldn't qualify it as a 'release' either).
Then I got myself a Sparcstation. Think of the light shining down from the heavens and illuminating the heathen's face for the first time. The hardware is beautifully built and high quality. The Linux support for it is excellent. As far as I'm concerned, the primary Linux platform is sparc; i386 is just a secondary little-used port. Linux on sparc is everything an OS should be - stable, fast, powerful, and elegant. And no matter how hard you try, you just can't do something this nice on shoddy, inferior hardware. Go buy yourself a cheap sparcstation 20 and find out what you've been missing. I've never tried Linux on Alpha, but the sparc port is proof that a port can be better than the original.
In comparitive terms, the Itanium is to the P6 core what the P6 core was to the original Pentium (P5) core. The Pentium brought ILP to the x86 line in a very big way and blew away the 486 with it's Superscalar architecture, and the P6 blows away the Pentium with it's speculative execution and other features, etc. The Itanium will be about as big a change. The disadvantage though is that to fully take advantage of the processor, programs and OS's need to be recompiled, though hopefully this won't be too big of a problem.
In fact, as early as August 31, 1999 -- when the ia64 chip was known as Merced instead of Itanium -- Microsoft announced that they had Windows running on engineering prototype systems. See also 64-bit Windows Overview at Microsoft's site, which notes that there is an existing Software Developers Kit and Device Driver Toolkit, specs for the Win64 API, etc.
Of course, clicking the link to the SDK and the DDK results in:
Let me tell you a torid tale of a consumer market gone bad. There was a company of name Microsoft, who's OS was pretty sad. It was a graphical shell of a crappy cmd line and it hadn't true multiuser support. And at Microsofts head was a man named Bill, who wrote BASIC all from scratch. And the DOJ got put away until Bill released 2k. When Windows 2k was not Ok. The DOJ got pissed. And they sued MSFT for half their cash and Bill got really miffed.
At the same time there was Linux, an OS that was pretty good. It was posix nice, with apache and Bryce, and was free and open as this. And Linux suits like RHAT sucked because they couldn't turn a profit: Who's going to buy a product they can download free from the 'net? And so the Linux hackers hacked away and made Linux for the I-64. And Bill got pissed and he was miffed and made a dangerous call.
His friends at Intel picked up and cut him some slack. They conspired to get the DOJ off his back. They slowed the I-64 so Bill could get in, and rewrote the specs so that Linux can't win. And Bill got good and Linus got pissed and the DOJ saved the day. Cause MSFT went down and LNUX went up and RHAT flew away.
That post was released under the GPL. Feel free to do whatever you want with it.
I don't have any references to back me up, but the sketchy reports I have read say that the long delays are because of horrible performance when executing legacy x86 code. Like Pentium II at 10 times the cost type of performance.
Sorry for the "I read somewhere" post.
-B
Breakthrough seems exaggerated.
The IA64 is not a breakthrough chips. There exist numerous other 64-bit risc chips already, with comparable performance (ultrasparc, soon new version, alpha, power). Also the IA64 is not fully backwards compatible with IA32 (finally), that is current IA32 binaries may run on it but at strongly reduced speed AFAIK (for those, upgrading to IA64 wouldn't make sense).
So I don't see what the fuzz w.r.t. the IA64 is all about. At the time it will actually hit the market, it is already outdated, and for example Linux will have faster (sparc/alpha) platforms it can run on.
Look your are going to cane me for this, but I did the install on an unsupported alpha. It works, and has worked for the past year and is working right now, providing me with squid proxy and ip masquarading but...
It does strange things, things that never happen on intel, it doesn't have the support infrastructure of intel. When you have a problem on a X86 machine you get onto deja.com and you find the answer, alphas are much more rare and it just takes longer to get answers - particularly as I live in Australia and most of the world is asleep when I am trying to do stuff (especially on unsupported variants).
Please understand that this is one of several alphas that we own, we run VMS and TruUnix and the alphas are _outstanding_ pieces of equipment. It is just more difficult to do linux on alpha than it is to on x86
My point (if I had one) was that ia64 is not for a casual user. There are significant differences and difficulties once you step outside of the most commonly installed base of equipment. If you install linux on a ia64, you will have a greater degree of difficulty compared to a ia32.
I _love_ linux cross platform capabilities, I use that ability, now if only all apps were cross-platform as well!
It amuses me when I write ia32. This name implicitly assumes there was a architectural basis for its design. This is of course entirely false.
Would you mind pointing out something that Windows does better? As for stability, efficiency, configurability and development-friendliness I think Unices are natural winners. Would you count wider software support as something an operating system "does better", or what do you mean?
A million flies can't be wrong... or could they?
I think, therefore thoughts exist. Ego is just an impression.
I saw 2 Itanium boxes at CeBIT. One was running Linux flawlessly. The other was running Windows. I brought up the control panel, and it crashed saying that it could find some DLL in WinNT32.
I have browsed across this link at www.faceintel.com. It really saddened me that such a respected and rich company such as Intel would give their employees a time in hell. Somehow, after reading the letters written by Intel employees, I see every effort on Intel's side with new eyes.
Maybe you will say it happens everywhere. I don't know, it doesn't happen in Nokia, that's for sure.
Now go on and moderate me -1 Offtopic, but tell me whether you can look at Itanium and say how cool it is after reading those letters.
Sigged!
Well, for starters, 64-bit systems usually comes with 64-bit data paths (buses). I'm saying usually because one could in theory implement a 8-bit bus that semantically behaved like a 64-bit bus. For instance, Amigas have 32-bit CPUs, but were equipped with a 16-bit bus. Having wider buses helps a great deal for most applications since the memory bandwidth is increased.
Also, 64-bit systems usually comes with a bus address path that is more than 32 bits. For instance, MIPS is 64-bit and is equipped with a bus that accepts 54 address bits (is this correct?). This enables you to put more than 4GB of memory into the machine and actually access the memory in a clean way from you appliction.
One of the most intriguing aspects of 64-bit architectures though, is that your virtual memory addess space is 64 bits. This allow you to implement things like single address space operating systems. You can also have huge files mapped into your address space (imagine having a movie of, say a few hundred gigs, mapped into your application), easing the development of applications tremendously. You could denote a few upper bits of an address to identifying hosts, and implement a large cluster of distributed shared memory. You could denote parts of the address to containing the process id, leaving you with a system where one particular address is unique both over time and among other applications. The list goes on...
As you can see, 64-bit architectures offer some obvious performance advantages. The biggest advantage though, as I see it, is that it opens up the door for some interseting operating system research/design.
but news.com has a big story on it right now - apparently HP is releasing an itanium emulator for people to use to test ia-64 code on, code compiled with SGI's OSS ia-64 compiler. :P
:P
apparently, there are only about 3,000 working ia-64 systems in the universe right now. (barring, of course, extra-terrestrial coincidental development of the ia-64)
--
blue, posting with mozilla m15
i browse at -1 because they're funnier than you are.
The entire Itanium platform is basically nothing more than a test platform for the next IA-64 chip, McKinley.
.25u process and in 1998), and appears to have only 96k L2 cache and 64k L1 cache, because rather than reaping the predicted simplifications of moving to a VLIW design, the Itanium core has fallen into a stew of complexity which leaves very little room for large cache size, even at .18u. This project has chewed up and spit out all of Intel's seasoned engineers on their Santa Clara design team, and was basically finished up by a huge committee of inexperienced kids fresh out of college. Every knowledgeble independent hardware analyst I've seen has said that, while IA-64 has some potentially promising ideas behind it, Itanium is going to be an ugly plodding beast of a chip, and its only chance at success is through miraculous marketing. Too bad for Intel's marketing machine that they're trying to move into a new space (high-end servers) with Itanium, and thus the people buying these machines are going to be well-informed and used to buying from vendors other than Intel.
.13u process, which ought to leave it enough room for a real L1 and L2 cache, and allow for presentable clock speeds. Thus, most organizations are looking at Itanium as a sort of public beta for McKinley--they'll buy a couple to get used to the platform, start to develop software for it, and have a core understanding in place in case they decide to upgrade all of their high-end servers to McKinley in a year or so. Almost no one is planning on deploying Itanium as a long term solution, because it will simply not be cost effective for the mediocre performance it will have.
This is because Merced/Itanium is 2.5 years late, and poorly designed. It is struggling to hit a meagre 800MHz (it was designed to debut at 800MHz, but on
The good news for IA-64 is that McKinley, the successor to Itanium, is much more on track (it should be out in volume by Q2 or Q3 2001), and appears to be performing quite nicely. Incidentally (or not), McKinley has been designed almost entirely by a much smaller (and more experienced) team over at HP (where EPIC, the philosophy behind IA-64, was designed as well); about all Intel will have to do with it is the fabbing and the Intel Inside logo. Also, McKinley will have the advantage of being fabbed with a brand-new
And thus the "race" to get a good IA-64 OS up and running really isn't all that important. MS will be sure to have W2K-64 "gold" well before McKinley is released, and that's all that really counts. Getting Linux on Itanium solid before W2K-64 goes gold will be nice, but it most certainly isn't going to win Linux a beachhead in the high-end corporate market, simply because most high-end corporate stuff can afford to wait for W2K-64 to be ready before they switch to IA-64.
On the other hand, a rather peculiar consequence of the IA-64 design is that applications need to be recompiled for each new processor core in order to take advantage of the EPIC features. (This is because, in IA-64, the compiler does most of the instruction scheduling, and not a scheduler in the CPU's logic; thus, the compiler better know how many functional units the CPU has, for example, or else it will schedule things all wrong.) This might seem to give an advantage to OSS, since one can just get a new compiler and recompile all of one's applications if one has the source to them. However, considering the market all of this is directed towards, I'm quite sure no one will have any trouble getting seperate versions of W2K or Oracle or whathaveyou optimized for Itanium or McKinley or any subsequent IA-64 chips.
For anyone who was expecting IA-64 to show up on the desktop anytime soon: don't. The first IA-64 chip scheduled to have a shot at the high-end consumer market is Deerfield, planned for 2004, but expected to come out later if at all. It would take a little bit to explain why, but EPIC is very much better suited to single-tasking machines running streamlined, well-tuned code--things like databases. It's just a poor fit for consumer applications. And anyways, if you want a neat Intel chip to get excited about, Willamette--the next desktop chip--looks way sexier than Itanium, and has a lot of cool design features (eg. double-pumped ALU, trace cache) which are just as innovative as IA-64's VLIW stuff.
Just to close things off, here's a couple links for those looking to learn more about Itanium and IA-64. (Even if the design doesn't pan out, some of the principles behind it are very interesting.)
Hannibal at Ars Technica on IA-64: a bit old, but a very well written overview of the design theory behind EPIC.
Hannibal on IA-64 vs. Sun's MAJC: mostly about MAJC (very interesting in its own right!), but a good example of a different approach to VLIW than IA-64's. (Crusoe is yet another example of a completely different approach to VLIW.)
Paul DeMone's Itanium article on realworldtech.com (unfortunately, the site seems to be down at the moment, so I can't get a link): an impressive technical argument about why Itanium might not actually achieve higher ILP in actual conditions, along with an interesting historical parallel of why this wouldn't be the first time Intel pushed a radical new chip architecture and it completely flopped (I forget the name of the previous chip, but it's an interesting story).
I doubt benchmarks from sources other then Intel will be legally available. I for one had to go through a twenty page NDA debriefing
"My opinions are my own, and I've got *lots* of them!"
Check out this article from Performance Computing. Describes the IA-64 and its advantages.
I agree that no one feature or 'thing' will make Linux the dominant OS. But one other thing to think about: Software is quickly getting more complex (and in MS's case, much larger), and the development times must suffer accordingly. On the other hand, new hardware is appearing faster (or so it seems). The question is, will any single company be able to keep up with getting their software onto new platforms quickly and reliably? MS is having problems, and they have mind-boggling amounts of resources to throw at it. Linux, on the other hand, gets ported quickly, and almost by habit to new platforms. IMO, the Open Source development model itself will eventually make Linux (etc), a more dominant OS, simply because it's far more responsive. Maybe that's the 'one thing'.
> Microsoft isn't as far behind as some would like to think.
What I want to know is, how far is the MS version away from going "gold", and more importantly, how much will it rely on the Itanium's 32-bit backward compatibility support for the next 3-4 releases? Windows does not have a good track record on 64-bit architectures.
> Rest not on your laurels, and keep up the good work.
Excellent advice. Still, this is something high-profile we can point to next time someone says "OSS is an imitator, always playing catch-up."
--
Sheesh, evil *and* a jerk. -- Jade
Actually, no, I'm not. I even agree with much of what you've said. When things do go wrong, the chances of finding someone else with the same problem (and hopefully a solution) are orders of magnitude smaller for non-x86 platforms.
It does strange things, things that never happen on intel
What I'm trying to find out is what strange things happen. As I said, I haven't noticed any on non-x86 Linux, and I'm curious about what differences you have found.
"The invisible and the non-existent look very much alike." -- Delos B. McKown
As noted on redhat's own news site, Microsoft has demonstrated pre-alpha versions of windows on ia64. The "Stunning news from MS on this front is long overdue." comment is somewhat off-base. Microsoft isn't as far behind as some would like to think. Rest not on your laurels, and keep up the good work.
I tried linux on a Digital Alpha and this _cured_ me of most of my desire to try alternative platforms. The trouble is that a whole lot of things just dont work on other than the X86 platform. Unless you can get a particular benefit from the chipset, for example the alphas fpu, I would suggest you stick with ia32.
Don't get me wrong, I like the multi-platform ability of linux and I think that the S390 and StrongARM ports are especially cool, I just wonder how sucessfull ia64 is going to be. Particularly against established platforms such as alpha and sparc
It will proberbly be huge, just to spite me, but let me just say that I wont be a early adopter!!
> It's really not the "64-bit" that makes it better / faster
However, if you have been doing scientific calculations using 64-bit numerics on a 32-bit machine, I would expect you to see very substantial performance strictly from the "64bitness" of the new system.
--
Sheesh, evil *and* a jerk. -- Jade
Don't forget about forestalling the Y2.038K 'bug' for another few billion years!
Eric
Theory two: I'm just starting to read The Tipping Point, by Malcolm Gladwell of The New Yorker. His thesis is that by applying theories from the field of epidemiology to other situations you can find explanations for seemingly inexplicable fads or trends. He argues that the same rules that apply to the spread of the flu apply to anything that is contagious. We all can attest to the contagious nature of Linux, so according to Gladwell it is simply a matter of reaching critical mass. He would say that there is a threshold, perhaps a percentage of users who switch to Linux, which acts as a "tipping point" where suddenly the balance is shifted and suddenly everyone else will switch. One of his examples in the book is fax machines, which came on the market in the seventies, but didn't reach critical mass until 1987 when enough people had them that it suddenly became something that everyone had to have. For cell phones, he says this happened in 1998, when the late adopters started buying them in droves. So we need to keep spreading the Linux germ, infecting our friends and acquaintances until we reach critical mass. It would be interesting to look at past shifts, like Wordperfect->MSWord or Lotus123->MSExcel, to see what percentage threshold was the tipping point. It may still be years away, but I like this theory, because it would be fun to see Linux flying off the shelves in glory as Windows drowns in its own $#!%.
"What I cannot create, I do not understand."
Since then I attended an SGI presentation about what they were doing to the IA64 GCC (Changes that will be or have been released to the Open Source community) and I changed my mind. Windows may be underwhelming on Itanium, but UNIX/Linux will be amazing. The timing couldn't be better, since we're just starting to see a corporate resurgance of UNIX, so I expect a lot of big companies to go to IA64 for server farms, web servers and the like.
I doubt it'll be as big a hit on the consumer level since the systems will probably end up being extremely expensive.
I'm trying to teach myself to set people on fire with my mind... Is it hot in here?
It includes a 64-bit server version of Windows 2000, a Microsoft C/C++ compiler and linker for IA-64, Intel C++ and FORTRAN 90 compilers for 64-bit Windows and 64-bit Windows 2000 libraries, header files and run-time libraries.
I supposed that I will be marked down as trolling. I'm not. I'm merely pointing out that this story was not properly researched before posting.
Linux is not the first to release a 64-bit Alpha release for the Itanium. The biggest difference here is that the Linux Alpha is available to the general public, while the Win64 version is only available to OEMs and key ISVs.
-jerdenn