IBM To Demo OpenPower 710 At SCALE 3x
An anonymous reader writes "IBM will demo their PPC based servers including their new OpenServer 710 at SCALE 3x this week. In addition they have their i5/520 running Power Linux, Intel Linux, AIX, i5/OS (OS/400), and Windows all simultaneously. SCALE will be held this weekend in Los Angeles at the LA Convention Center. Speakers include Kevin Foreman (Real Networks), Jon Hall (Linux International), Larry McVoy (CEO BitMover), Marc Hamilton (Sun) & 30 other sessions. In addtion to the talks there will be over 40 exhibitors including IBM & Novell. If you're in LA drop by on 2/12-2/13. There will also be a dinner and GPG Key Signing party. (For a free exhibit hall pass register with the promo code "FREE" or a discounted full access pass with "NEWSP".")
I think they should use the distributed.net client for benchamarking. maybe they can take a few decades off of the RC5-72 challenge, or finish up OGR. :)
Pretty Pictures!
The sims was ported to Linux using winelib, so you don't need winex or the windowsversion of seems to play.
Marc Hamilton (Sun)
I guess Luke is Sun's only hope?
I seriously doubt the ability of your 3 yr old pc to run all five operating systems simultaneously. It might be able to into any one of them at different times but any processor you might have would be unable to handle multiple OSs.
how many linuxes does it run?
a beowulf cluster of linuxes, all of them running on the same machine.
Register before Feb 6th for a chance to win a copy of "The Sims" for Linux
Now you tell me! You twat.
The anonymous reader doesn't bother to state where that info came from. I would assume to be a member of the group putting on the expo/convention.
Hi Luke.
Are they running it under emulations or is there once again a PPC version of windows?
See my blog http://ilovecookes.blogspot.com/ for light hearted technical information.
$30K seems a bit pricey, which model # did the IBM guy quote for u?
h ar dware/710_browse.html
http://www-1.ibm.com/servers/eserver/openpower/
9123-710A POWER5 / 1.65GHz 1-way 36MB 2048MB 2 x 73.4GB Ultra320 10K rpm $4,713.00 IBM Web price*
9123-710B POWER5 / 1.65GHz 2-way 36MB 4096MB 4 x 73.4GB Ultra320 10K rpm $8,428.00 IBM Web price*
9123-710C POWER5 / 1.65GHz 2-way 36MB 8192MB 2 x 73.4GB Ultra320 10K rpm $12,766.00 IBM Web price*
Maybe you got quoted a different model? The 720 model is made to fit 64GB of RAM, and fits four CPUs.
From The Register:
Big Blue bills its OpenPower line as a serious threat to Unix gear from the likes of HP and Sun Microsystems. But at a starting price of $3,449 the OpenPower 710 will also rival systems running on Intel and AMD processors.
HA!
Look at the prices for _good_ SCSI hard drives, and already your close to $1000. Segate 75 gigabyte 10k disks are around $300. $600 for 140ish gigabytes.
Next thing you've got to look at is processor cache. It's expensive anywhere you put it, so most low-end Intel processors have around 1 to 2 megabytes of it. A 30 grand OpenPower machine is likely to have double or triple that (he said not knowing).
Plus a lot of other things I don't feel like rattling off right now.
A $1000 OpenPower server would be nice, though. So would a Cray that fits in my pocket that costs $25.
To run Intel Linux / Windows on i5 (ex AS400/iseries) you need to purpose either card with intel cpu/memory (like blade) or special kit which can be used to connect i5 box to the intel (xseries) server. The main benefit of running intel apps on i5 is single-level storage. :)
Anyway it is a great box - i'm happy having one near my desk for two weeks. It is black, heavy, and rather quiet (for server)
If they would make more money at a lower price then they probably would sell them cheaper. Lets just say an openpower server costs them $1000 to build. If they sold 100 of them for $10,000 they make $900,000. If they sell them for $2000 they would need to sell 900% more of them while competing with a million other low cost, somewhat lower preformance, server solutions. If they sell them for $10,000 then anyone to whom it's worth paying for the extra preformance in saved operating costs will pay alot more for that somewhat better preformance, whether they charge $20,000 or $10,000, so might as well charge them 20.
I for one welcome our new vengeful sith overlords.
Others have mentioned the prices go lower, but additionally, if you have a blade infrastructure or are building a large scale network and can benefit from blade infrastructure:s /stores/servlet/C ategoryDisplay?categoryId=2586156&storeId=1&catalo gId=-840&langId=-1
http://www-132.ibm.com/webapp/wc
A Power server that is actually priced competitively with the Intel blade servers. True the bladecenter
chassis isn't well suited to some environments, but there is an option.
Oh, and there is always Apple...
XML is like violence. If it doesn't solve the problem, use more.
He's just A New Hope.
put the what in the where?
It's way more fun and easier to use the PPC instruction set than the x86's.
Sadly, it costs $40 plus shipping on ebay, but it'll come down soon.
Not relevant, but highly amusing. And yes, I know the C1 has archetechtural advantages over the Zire 21 (parrelelism, floating point...). It'll happen to the OpenPower servers too.
Sig:Why copyright isn't a fundamental human right
So... Basicly what you are saying in this post is that cutting the price down to 3% of what they are currently asking would make more people buy the machines? Pure genius.
Why? x86-64 does pretty well in benchmarks, and it's cheap as hell.
Even if you pick your chips for assembler elegance, x86-64 isn't too bad. Most of the x86isms that hampered performance are either not used anymore (e.g. segmentation) have been fixed to some extent or no longer matter much in a modern chip running modern code.
In the "fixed to some extent" I'd include the lack of integer registers - x86-64 has twice as many, the nasty FPU architecture - x86-64 uses SSE2 instead. Also, variable sized instructions can be executed efficiently if you transform them into uops and give a higher code density than a typical Risc, so you get better cache hit rates for a given cache size.
And by "modern code on a modern chip" I mean that if you look at the code that comes out of a modern compiler, it's almost as good you would have got on a clean Risc design, but you run it on a chip with a clock rate that is much faster than Risc chips manage, because the x86-64 world now has two companies competing, and they can both afford high end fab plants.
In fact I'd say that x86-64 will eventually kill all the Risc chips, PPC included in the desktop/server world. The embedded world is different of course.
echo -e 'global _start\n _start:\n mov eax, 2\n int 80h\n jmp _start' > a.asm; nasm a.asm -f elf; ld a.o -o a;
Yeah... for all the people who write assembly these days....
I've written PPC assembly but no x86. I've never been a fan of x86 but these days, so few people program in assembly it's not an issue.
For subset of users it is a useful or valuable quality. btw, Some people think that programmers should know at least 1 assembly language and the PPC is a good cpu to learn it.
"but you run it on a chip with a clock rate that is much faster than Risc chips manage,"
My PowerMac G5 runs at 2.5Ghz - how fast do Opterons scale up to now?
That was classic intercourse!
Double or tripple that for $30K?
More like 72x that for $23K. IBM doesn't mess arround when it comes to cache.
Jon "maddog" Hall was one of the first supporters of linux way back in the early 90's. If I remember correctly he was the one who donated a DEC Alpha(I think he worked for DEC) for Linux's first port. I may be wrong...
I just read this entire thing at -1, and there hasn't been one post that indicates anyone has even seen an OpenPower box. We purchased an OpenPower 720 box a few months ago. 2x 1.65 GHz procs, 4 GB RAM, 2 73 GB drives, 6 146 GB drives for about $22,000. We run SLES 9 on it, and use it for ITSM. I chose it opver and x86 box for one reason: I/O, which is pretty much where all big-iron architectures trounce x86. There have definitely been growing pains, though. I had to work with SUSE on a fix for installation onto an IBM RAID array. Also, I cannot utilize all of the disk space that was purchased. Both of those issues are results of IBM controllers requiring you to format a drive with 522 byte sectors. The results are workable and perform well, but not quite the "certified" solution I was expecting. Anyways, my 2 cents. I still think I made a good choice, considering management wouldn't go for a PSeries/AIX box.
http://www.tbreak.com/reviews/article.php?id=335
Athlon FX-55 - 2.6GHz.
I would say thats faster than a G5. AMD and IBM have been sharing alot of processor technology lately, its not suprising that there processors run about the same speed. more info
If only I had some mod points today I'd mod you up for sure. We just developed a custom app with about 3000 hours at $125 an hour. By my math that is $375,000. Do you really think they care if the server is $1000 or $10,000 if the $10,000 is supported better? No, of course they don't. This guy is more with it than 90% of the slashdot posts that I read.
I agree... it was actually a part of our class schedule in college. In high school, I tought myself 6502 assembly on an Apple ][+ and I learned M68000 family and MC6800/6809/6811 assembly some on my own but some for classes. Later, I dallied in PPC and SPARC and a touch of MIPS. I think many programmers would benefit from retaking data structures in assembly :)
However, the fact is that there isn't that much call for assembly outside of embedded systems these days.
PS: I'm trying to decide whether or not I want to learn x86-64 assembly. I'm afraid it might be too much like x86 which I think is just ugly, but I haven't looked at it enough to decide yet.
One of the really fascinating features of this server is that uses IBM DLAR (Dynamic logical partitioning). This means that new instances of any of the supported OS's can be instantiated or removed as needed.
your pc can run OS/400 and AIX? roflmao
I think they meant OpenPower............
These IBM OpenPower machines are MUCH more powerfull then a Opteron, or a Apple g5.
0 00325
They are NOT Power970 proccessors used in Apple machines, these are POWER 5 proccessors used in Unix big Iron. Opterons aren't even in the same catagory.
They get spanked by quite a large margin in Opteron vs POWER. Comparing a Opteron vs a POWER 5 is like comparing a Intel Xscale proccessor vs a Pentium 4 "Extreme".
The transistor counts are astronomical in the POWER vs Opteron for one thing.
See here:
http://www.aceshardware.com/read.jsp?id=65
Notice how the low-end Power 5 proccessor dominates the top of the line Opteron in all catagories.
In Specfp2000 you have a single proccessor vs single proccessor. A 1.65ghz Power5 gets 2138 PEAK score while the 2.2ghz Opteron gets 1691
In the 4 proccessor setup a 1.9ghz Power5 proccessor scores double in floating point performance then what is possible with 4 2.4ghz Opterons.
Plus the Power 5 proccessor scales up to 32 in standard configuration, while the Opteron is limited to 4 cpus.
There is no comparision between the proccessors. The Power5 dominates every single catagory.
Including heat, power requirements, and expense.
However if you notice that the prices for Opteron Sun Solaris hardware falls with in the range of pricing for IBM's OpenPower servers.
Seems like IBM is going to use dirt cheap prices for Power to run Sun's opteron setup out of town.
Also, unfortunately for Sun.. higher end POWER setups dominate their nicest Sparc machines, also.
Maybe Timothy is smelling SCO here.
No shit!
2 CPU cores per module, 6MB? L1 cache and 36MB L2 cache per core, all on one P3 sized module.
John