Posts by Kebabbert

IBM heaves new System z minis at mainframe shops

@Oolons

"...Do you not engage your brain before typing out a load of daft anti-IBM propaganda? If even a tiny amount of what is in your post was true then do you not think everyone would move their workloads to x86 servers?..."

Again, let me repeat, I am only saying that Mainframe cpus are very weak. The evidence are numerous, as I have provided several links from Mainframe experts. If you have written an Mainframe emulator, then you must be considered an "expert", yes?

Because I have provided lot of links and proofs, my post can not be considered as "propaganda". In fact, my links are true. I am not trying to spread false rumours, making unsubstatianted claims. What I say is true, there is major substance in my claims. The IBM Mainframe cpus are very weak. Just look at all the links I provided.

If you have counter proofs, then I will gladly read them. If you can show benchmarks that show how fast Mainframes are, then I will of course stop say that Mainframe cpus are weak. But, there are no such benchmarks. Why? Maybe there is a reason that IBM never publishes Mainframe benchmarks? Have you thought about that? Maybe the links I collected here, are correct? Maybe you are fooled?

.

.

"...As an aside in the real world (Not a place for a Kebbabert to tread) a large banking group..."

I don't work in the banking world. That world is boring. Come on, Mainframes?! How sexy is COBOL? Who wants to work with antique and slow hardware such as Mainframes? What does a mainframe do? It updates lot of accounts with a % batchwise. Do a trivial calculation on lot of bank accounts. How sexy is that?

I work in finance. Extremely low latency, very high throughput, extreme cpu demands, algorithmic trading, High frequency trading, risk calculations, etc. Now, that is sexy! Extreme tech. Lastly I heard about HFT traders using FPGA to implement their algorithms! Apparently FPGAs can lower the latency 10x or so. That is cool and exciting indeed. Forefront tech. A Mainframe is not used for such things, and could never do that. Too slow, and bad latency.

There is a reason stock exchange systems are not run on Mainframes. If a exchange system crashes, the exchange will loose lot of money, and customers (hedge funds, IB, etc) will start to trade on other exchanges (they can not sell/buy when they see an opportunity and loose lot of money, often they leverage 5x or more). If Mainframes were fast, and reliable (I have several links that show Mainframes to crash, one reason the software emulator TurboHercules is sold is because customers need failover when their Mainframe crashes) then the finance world would not care how much a Mainframe costed. The Wall Street firms would get Mainframes no matter the price. But guess what? Wall street firms are not using Mainframes, they use Unix and Linux. Why is that? Maybe you should start ask you some questions?

Only the slow banking world, without the need for extreme performance use IBM Mainframes. Finance does not.

@Jesper Frimann

"...You didn't really read and understand what I wrote now did you ?..."

And I must ask you the same thing: you didnt really understood what I wrote, did you? I am ONLY claiming that Mainframe cpus are weak, and have provided numerous evidence for that. That is all I claim.

You mock me for believing an x86 server could replace a Mainframe. No, that is not what I am saying. There are I/O intensive workloads where an x86 can not compete with a Mainframe, in that case you can not replace a Mainframe. Why are you trying to make me sound as if I claim a x86 can replace any Mainframe work load? To make me look dumb?

Let me repeat, in case you did not understand, the Mainframe cpus are weak. The evidence are numerous. If your Mainframe workload is mainly cpu intensive, then maybe you can replace it with x86 running an software emulator. That is all I say. Please stop making me say things that I have never said.

@Jesper Frimann

I am focusing on CPU performance in the post above, not I/O. So, when I say that you can download TurboHercules and use it to virtualize Mainframes, I speak about CPU performance. An x86 server would never be able to handle the I/O of a Mainframe. I have never questioned a Mainframe's ability to do superior I/O. I have heard that a big Mainframe has 296.000 I/O channels.

If you focus on Mainframe CPU performance, then you can emulate the work load on an x86 server.

If you focus on Mainframe I/O, then you can not emulate it on an x86 server. Then you need to stay on Mainframes.

.

But traditionally, Mainframes are used in banking: Update large number of accounts with a simple calculation (increase interest rate). Lot of accounts involved. Batch work. Much I/O. Cpu performance is not the focus.

In finance and in stock exchange systems, you typically use much cpu performance. And also I/O of course. But larger systems are distributed so there is no single server that has to handle lots of I/O. For instance, risk calculations take much cpu. And also MCMC methods.

@Paul Crawford

You can not just use IBM Mainframe MIPS to compare against normal x86 MIPS. So that is not really comparable. Also, IBM marketing will try to make it difficult to compare performance between x86 and Mainframes. For instance, it would be easy to run Dhrystones / SPECint / SPECfloat / any other benchmark on a Mainframe, but you will never see such benchmarks published. IBM marketing would never allow publication. But we know that whenever IBM has anything advantageous to say, IBM says so. So, why are IBM not publishing common benchmarks run on IBM Mainframes? IBM Mainframes crush everything in terms of performance, right?

.

The new IBM Mainframe Z196 cpu, is the "Worlds fastest cpu" according to IBM:

http://www-03.ibm.com/press/us/en/pressrelease/32414.wss

It has crazy specs. 5.26GHz, ~300MB cache (L1+L2+L3). Close to half a GB of cache! So, how fast is it? Let us see. It is 50% faster than the previous Z10 cpu. How fast is a Z10 Mainframe cpu? The biggest Z10 Mainframe with 64 cpus, give you 28.000MIPS. Now, that is a lot! How much is it?

.

If you use an 8-socket Intel Nehalem-EX server, and use the IBM Mainframe software emulator "TurboHercules", then you get 3.200MIPS:

http://en.wikipedia.org/wiki/Hercules_%28emulator%29#Performance

That is under software emulation, which is 5-10x slower than running native code. So, an 8-socket Nehalem-EX server would actually give 5-10x higher performance if you ported the Mainframe applications, so we could run native code. That is: 16.000-32.000 MIPS. Thus, an old 8-socket x86 server would give you as many MIPS as the biggest Z10 IBM Mainframe with 64 cpus. If you build an Emulator for IBM Mainframes, then you know the ins and outs of Mainframes.

.

Another developer that ported Linux to IBM Mainframes came up with this rule:

1 IBM Mainframe MIPS == 4 MHz x86.

http://www.mail-archive.com/[e-mail]/msg18587.html

So, the biggest IBM Mainframe Z10 with 64 cpus, correspond to 28.000MIPS == 112.000MHz x86. But, a single 8-core Intel Nehalem-EX running at 2.5GHz, has 20.000MHz in total. Thus, you would just need a few Nehalem-EX to reach 112GHz. Again, we see that you need a few Nehalem-EX to beat 64cpus of the Z10 Mainframe.

.

And, Here is a source from Microsoft

http://www.microsoft.com/presspass/features/2003/sep03/09-15LinuxStudies.mspx?

"we found that each [z9] mainframe CPU performed 14 percent less work than one [single core] 900 MHz Intel Xeon processor running Windows Server 2003."

The z10 is 50% faster than z9, and the z196 is 50% faster than z10, which means a z196 is 1.5 x 1.5 = 2.25 times faster than a z9. This means a z196 corresponds to 2.25 x 900MHz = 2 GHz Intel single core Xeon. But todays modern server x86 cpus have 8 cores, which means they have in total 8 cores x 2 GHz = 16 GHz. Again we see that you need only a few 8-core Nehalem-EX to match the biggest z10 Mainframe.

.

Now the Z196 cpu is 50% faster than the z10 cpu. Instead of using 8-socket x86 servers, you need to use 12-socket x86 servers to beat the biggest Z196 IBM Mainframe, utilizing the "worlds fastest cpu". In short, if you can migrate your workload to x86, then do it.

.

.

Sure, IBM claims their Mainframe can virtualize 1.500 x86 servers. If you read the entire study, you will see that IBM assumes all x86 servers to idle, and the Mainframe is 100% loaded! Now what happens when a powerful x86 server does some work? The weak Mainframe can not virtualize that load. The Mainframe cpus are to weak.

.

Sure, IBM claims the Mainframes have brutal I/O and can handle 400.000 logged in users. But I would not be surprised if 95% of the users idle and only a few of the users do some work.

Unix still data center darling, says survey

@Jesper Frimann

"...Ehh ? What the f word are you talking about. The POWER server platform actually has good scaling. Well at least compared to anything Oracle can muster..."

Sure, you talk about one benchmark; SPECint. And you prove that POWER scales well on SPECint, which is an easy parallellisable benchmark. Does this mean you have proved that POWER scales well in general? I dont agree with you:

I read here on this site, that AIX needed to be rewritten to handle P795, with 256 cores. So IBM could not handle 256 cores until just recently? Isn't that bad scaling?

I also heard that IBM can not scale the TPC-C clusters to counter Oracle's TPC-C world record. Isn't that bad scaling? IBM will never be able to break Oracle's TPC-C record?

I also heard that IBM's biggest mainframes only has 24 cpus. Why not bigger? Problem with scaling? I just asking.

If we talk about Solaris 11, it has been rewritten to handle big Oracle servers with 16.384 threads. Even the old Solaris 10 handled 256 threads. Sun sold old SPARC servers with up to 144 cpus.

.

.

"...You do know that POWER have managed to do all three things. Put more cores on a chip AND increase the per core throughput and socket throughput..."

No, that is not that I am talking about. I am not talking about if IBM increased throughput and needed to lower GHz to stay in a reasonable wattage.

I am refering to when IBM explained that the future is in 1-2 super fast cores beyond 5GHz and higher, because databases prefers on strong cores with good single thread performance. When I look at POWER7, I dont see 1-2 cores clocked higher than POWER6, at 6GHz or 7GHz. Instead, I see many cores, lower clocked than POWER6, going under 5GHz. I wonder if POWER8 will have more cores than POWER7, and even lower GHz? And thus, stray even further from the "1 super fast core at 6-7GHz"? Dont you agree that IBM has abandoned the "1-2 superfast cores" and followed Sun's "many lower clocked cores"?

"...Again you have absolutely no clue what so ever..." - does this mean you think that POWER7 is more similar to a single core cpu at 6-7GHz, than a cpu with many cores at 3-4GHz?

.

.

"...I think the guy that put this best was Linus, when he asked why does a filesystem have to do that ?... And cool as it is.. I have to say I agree with Linus, this is perhaps taking the role of the filesystem one step to far..."

I certainly dont agree with you. As ZFS creator Jeff Bonwick explained:

"The job of any filesystem boils down to this: when asked to read a block, it should return the same data that was previously written to that block. If it can't do that -- because the disk is offline or the data has been damaged or tampered with -- it should detect this and return an error...Incredibly, most filesystems fail this test. They depend on the underlying hardware to detect and report errors. If a disk simply returns bad data, the average filesystem won't even detect it."

I also know that several large instituitions such as physics centre CERN (who stores large amounts of data for their big Hadron LHC collider) are very concerned with this. If CERN stores experiment data and if the data is silently corrupted, maybe CERN will not detect the Higgs boson. You know, there are thousands of researchers spending years on this project. That is the reason CERN is very concerned with silent corruption:

http://indico.cern.ch/getFile.py/access?contribId=3&sessionId=0&resId=1&materialId=paper&confId=13797

Or, if you encounter a silent corruption in your database. When did the corruption take place? How long will go for backups? Half a year? One year? Database admin talks about silent corruption

http://jforonda.blogspot.com/2006/06/silent-data-corruption.html

Or in this case: a flaky switch is corrupting the data. ZFS was the first one to notice, because ZFS protects it's data.

http://jforonda.blogspot.com/2007/01/faulty-fc-port-meets-zfs.html

"As it turns out our trusted SAN was silently corrupting data due to a bad/flaky FC port in the switch. DMX3500 faithfully wrote the bad data and returned normal ACKs back to the server, thus all our servers reported no storage problems."

If you and Linus don't agree that the stored data should be intact, then you can not really trust your data, I hope you realize this? ZFS does what ECC RAM does: protects your data against power spikes, hardware problems, etc. I really do hope you have ECC RAM in your servers, but maybe you think ECC RAM is not necessary, just as you think that protective filesystems are not necessary?

Modern Enterprise SAS disks has 1 irrecoverable error in every 10^16 bits, just look at the spec sheet. Protection such as ZFS is necessary, in my opinion. But of course, you and Linus may have differing opinions, that is fine with me. But I would be careful, and suggest you read more about ECC RAM and Silent corruption. The study by CERN above is a good place to start. If you want, I have plenty of research papers on this. Just ask me if you want to start worry about protecting your data. Here is one link on ECC RAM, in case you are not familiar with ECC RAM:

http://en.wikipedia.org/wiki/Dynamic_random_access_memory#Errors_and_error_correction

To exemplify it's importance, for instance, Microsoft found out that many of the Windows crashes depended on non ECC RAM, which is why MS wanted everyone to use ECC RAM when running Windows. So yes, ECC RAM is important. Read the above link.

.

.

"...As for putting computers into a container... well... I would hardly call that innovation. We have them.. it's stupid in most cases IMHO, but as a hack to have variable capacity that you can move from location to location it's ok..."

I am just pointing out yet another case where IBM copied Sun/Oracle. And the BlackBox has its use, which in that case the blackbox is perfect.

.

.

Kebabbert: "What techniques has IBM created, that Solaris copied? You talk about "recent years". Can you give an example?"

Jesper Frimann: "...You have to be kidding ?..."

No, I am not kidding. Let me repeat my question: In RECENT years, what has Solaris copied? I know that IBM did great things in the 1960's etc. But in recent years? To me it seems that IBM is copying from others, but maybe you have some counter examples?

@Peter Gathercole

"...But in recent years, I have perceived [Solaris/HP-UX] to be less innovative than the IBM offering..."

I certainly dont agree with you. I mean, IBM offerings has scaling problems and does not scale as well as Solaris (TPC-C, AIX scaling was rewritten to handle P795 with a measly 256 cores).

IBM AIX is copying Solaris DTrace and renaming it Probevue.

IBM for many years trash talked Sun Niagara and said that 1-2 cores at high clock speed such as 5-6GHz is the future, because data bases like strong cores. To use many cores at lower speeds is just a bad idea, said IBM. One strong core is the future. And now today POWER7 does not have 1 core at 8-9GHz, but instead it has many lower clocked cores, just like Niagara. Sun realized that GHz race will shift to many core race, but IBM did not understand that until POWER7. POWER6 was 5GHz and 2 cores. POWER7 is not 6-7GHz and 1-2 cores. So, the future is not 1-2 strong cores. Back in the Sun days, 8 cores in a cpu was just crazy, no one had that, except Sun. Today Oracle aim for 512 threads in one cpu, which is crazy today. But tomorrow everyone will have it. IBM will copy that many threads too.

And for instance Solaris ZFS, I dont know of any IBM storage solution that protects your data as well as ZFS does. It would not surprise me if IBM copy ZFS too, soon.

Sun first released their Container full of servers, called Black Box. And some time later, IBM also started to offer a container full of servers.

etc etc etc.

What techniques has IBM created, that Solaris copied? You talk about "recent years". Can you give an example?

.

.

"...But I would say that AIX looks destined to the the last Genetic UNIX standing, given HP and Oracle's current attitude towards their products, and Linux still has a way to go in enterprise environments to replace it...."

So what do you say about official IBM statements to media and Linux conferences, that AIX will be killed off and replaced with Linux, some time in the future? I mean, HP-UX will still support Itanium 10 years from now even if Itanium is killed today. So in the enterprise setting we talk about 10 year time horizons. Not next year.

http://news.cnet.com/2100-1001-982512.html

What do you say about the trend of x86 cpus catching up on POWER performance? I mean, POWER6 servers were several times faster than x86, and costed 5-10x more.

POWER7 is 10% faster than Intel Westmere-EX and costs 3x more:

http://www.anandtech.com/show/4285/westmereex-intels-flagship-benchmarked

The next year Ivy Bridge will be 40% faster than Westmere-EX, according to official Intel statements. x86 is getting high performance, at a higher rate than POWER.

Does this mean that POWER8 will be slower than x86? In that case, POWER8 need to be real cheap. And we all know that IBM only does high margin business. IBM will kill off POWER it is too cheap, and will replace everything with fast and cheap x86 running Linux. Coincidentally, IBM has officially confirmed this: "AIX to be replaced with Linux". See post above. This is not something I make up, this is true. IBM has officially said this. I am not spreading false rumours, this is true.

@A.C

I remember some CEO of a big firm that replaced Solaris with Linux, and claimed that the Linux solution is so much faster. That was great, until I found out that he had replaced 800 old SPARC servers, with 4.000 new x86 servers with double Intel Duo Core cpus running Linux.

Do you think it would be fair if I compared Linux on an 800MHz SPARC server, to Solaris running on a x86 server with dual Intel Core Duos at 2.4GHz?

There are many cases where Solaris gives higher performance than Linux, on the same hardware. For instance, official SAP benchmarks.

@JEDIDIAH

"...Sun was the reference platform for Oracle for the longest time. Then it became Linux...It's the top Oracle platform for crying out loud."

No, this is not really correct. Solaris is the most common platform that Oracle is running on, according to Larry Ellisson, and he should know. Also, the "reference platform" to run Oracle DB on, is Solaris. It has been for a long time before Oracle bought Sun, and continue even today.

Intel 320 SSD bug causes forum despair

@AndrueC - NTFS is not safe.

Actually, I would be careful trusting NTFS as it is quite unsafe. Research shows that NTFS, XFS, JFS, ReiserFS, etc are all not safe.

Have you heard about ECC RAM? Why do you use it? Well, there might be spontaneous flipping bits in RAM. ECC corrects those flips. This problem also exists in filesystems:

http://www.zdnet.com/blog/storage/how-microsoft-puts-your-data-at-risk/169

And also, research shows that hardware raid has the same problems. For instance, there are papers from CERN confirming this.

Here is more reading if your data is important to you, you should read it. Here are lots of research papers:

http://en.wikipedia.org/wiki/ZFS#Data_Integrity

IBM preps mini-mainframe for launch

Software Emulator?

"...If IBM keeps the zEnterprise BC midrange mainframe at five CPs, then a single system image should be able to span up to 3,400 MIPS or so...."

Is there really a point of selling those midrange IBM Mainframes? I mean, an 8-socket Intel Nehalem-EX server gives 3.200 MIPS using software emulation with the Mainframe Emulator TurboHercules:

http://en.wikipedia.org/wiki/Hercules_%28emulator%29#Performance

IBM gets fat on Oracle-HP Itanium spat

@A.C

Well, as far as I know, there are other vendors developing and selling SPARC cpus. Doesnt this make it more open than POWER?

And SPARC has been officially open sourced. You can download the spec, and manufacture a SPARC yourself. Doesnt this make SPARC more open than POWER? Or?

@Lennart

Ok, I did not know this. Thank you for explaining this to me.

A question, it seems only IBM makes POWER cpus? Freescale does PowerPC cpus which is another thing?

http://en.wikipedia.org/wiki/Power_Architecture#Glossary

So, is it only IBM that does POWER cpus? Is this true?

Que?

"...I think that the reason we won't end up in the same boat as Itanium is that the Power market won't end up in the same spot," says Gelardi. "There is no software company in the world that wants to be dependent on one hardware supplier..."

What does he mean? Isn't POWER cpus only from one vendor? Just like Itanium?

SPARC is open sourced, so anyone can manufacture and sell SPARC. In fact, the chinese super computer on TOP500, Tianhe, is a SPARC derivative. And also Fujitsu sells SPARC.

Oracle Solaris 11 to abandon elderly servers

@A.C

"...Google AIX killed and you find only Kebabbert, the Swede,many, many times all over the Internet..."

I am flattered that you think of me and of what I write. Do you often think of me? How cute.

.

.

"...Classic FUD with the exception that FUD usually comes from sources with at least some credibility...."

:o)

How ironic that you call me a FUDer. You apparently dont know what FUD is. I suggest you read on FUD. Here we see that IBM was the first company to employ FUD. Yes, IBM, the same company you support. IBM supporters are well known for their FUD.

http://en.wikipedia.org/wiki/Fear,_uncertainty_and_doubt

And you, an IBM supporter, call me a FUDer? Well, when you FUD, you are basically, lying and spreading false rumours. For instance, "SPARC will be killed said Oracle executives, migrate to POWER now or you will loose lot of money". So, where have I lied? Is it not true that IBM officially said that AIX will be killed and replaced with Linux? Is this a lie? No, it is not a lie. I speak true, and you are trying to discredit me by calling me a liar and FUDer. This makes you the FUDer, yes? I speak true, not lies.

It is not my fault that IBM wants to kill off AIX, is it? If IBM never said that, I would not write that. I dont want to write things that I can not prove with links. Why would IBM ever say that?

.

.

"...- It is not an official statement..."

http://news.cnet.com/2100-1001-982512.html

I consider this as official statement: to the IT-media, and to LinuxWorld. How can it be more official??

"Steve Mills, senior vice president of IBM's Software Group, told CNET News.com at last week's LinuxWorld trade show. "

.

"...- It does not say that AIX will be killed..."

Well, if IBM is going to replace AIX with Linux, it means AIX must be killed, yes?

"Asked whether IBM's eventual goal is to replace AIX with Linux, Mills responded, "It's fairly obvious we're fine with that idea...It's the logical successor."

A replacement "won't happen overnight," Mills said, but years of experience designing operating systems at IBM and other companies means developers know just where Linux needs to go. "The road map is clear. It's an eight-lane highway." "

.

"...- AIX on Power gaining market share..."

So what? POWER6 servers costed 5-10x more than x86. POWER7 costs 3x more. POWER8 will probably be cheaper still. And we all know that IBM does high margin business and walk away from low margin. What does the trend tell you? Do you think POWER8 will be even cheaper than POWER7?

AIX might be gaining market share, but each POWER7 server costs much less than POWER6 servers. Solaris servers have always sold in much larger quantities than POWER, but each Solaris server has lower price.

.

PS. A hint, stop FUD and spread false rumours about Oracle and Solaris, and I will stop post this article over again and again. Apparently you IBM supporters are very uncortomble reading this article where it says that AIX will soon be killed.

But if the things you say about Oracle are true, negative or good, and if you can prove your claims with links, then you are not FUDing, and you are not spreading false rumours. And then I will be quiet. True rumours are not FUD, no matter if they are negative or positive. Just re-read the wikipedia article on FUD and you will understand. If you say true things, I am silent. I dont care if they are negative or not, as long as they are true.

@M1cha3l

As you know, I work in a large finance company. It is written in the article you posted. I do not work for Oracle, as you know. What are you trying to do? Discredit me?

Regarding my postings here, yes, I do post here. So what? There are lot of IBM supporters FUDing a lot, and spreading false rumours here. Why dont you remark on their behavior?

At least I am not spreading false rumours, I do not FUD as many IBM supporters here do. I can link to articles proving my claims, because I speak true things. IBM has officially said that AIX will be replaced by Linux. This is true, and not a false rumour.

Why dont you object when IBM supporters spread FUD here? Why do you object when I write true things with links to my claims?

@Oliver Jones

I would be cautious to move to AIX, as IBM officially said that AIX will be killed off, and replaced by Linux:

http://news.cnet.com/2100-1001-982512.html

However, Illumos is the open sourced version of Solaris and the community drives it. Several of the famous Solaris developers has quit Oracle and joined Illumos (guys behind DTrace, ZFS, etc). Open

Also, the reason that Oracle stops supporting the old SPARC cpus, is because Solaris 11 has a new killer feature that will be announced soon.

Xeon E7 servers run with the big dogs

@cmaurand

All cpus are getting RISC like. Even x86 has RISC similar architecture inside the chip. There is virtually no difference between CISC and RISC nowadays. They are all same.

@Beachrider

What 10% number is wrong? What do you mean? Is the highly clocked POWER7 not 10% faster in some benchmarks? No, the number 10% "aint right"?

So, the correct number is 11.5% in one benchmark, which the article stated. But how do you know POWER7 is exactly 11.5% faster than Intel Westmere-EX? Maybe the correct number is 11.4999%? Are you going to remark on that too? Have you checked it up whether it is 11.5% or 11.4999%? Why dont you check it up, and complain about that rounding too?

No, the point is, that POWER7 is only ~10% faster in some benchmarks. And Ivy Bridge will be 40% faster using the same number of cores - according to Intel.

@Splitbrain

You can run Solaris on x86 if you need a 32 socket Westmere-EX box. Solaris is the only Enterprise Unix (except BSD) that runs on x86. I personally prefer Solaris on x86, over Linux. But YMMV.

I agree that this Westmere-EX is a fast chip, the mighty POWER7 is only ~10% faster in some benchmarks. But wait until next year, when the Ivy Bridge based cpus arrive. They will be 40% faster, according to Intel. This means the fastest cpu on the market, will be Intel Ivy Bridge. And later, AMD will release it's 20 core Bulldozer cpus. And Solaris runs on them all. :o)

They shoot mainframes, don't they?

@zmarcel

"....I am sorry to see that the 20 year old version of; "my IBM AT server is as fast as a Mainframe" is now replaced by "My laptop can run 10 Mainframes". Often used by those who have absolutely no understanding of how Mainframes work...."

Actually, I am not saying this. I totally agree with your criticism. It is wrong to claim that "I can virtualize 10 IBM mainframes on my laptop" - I do not say that. Just because I can in fact boot 10 mainframes on my laptop, provided all idle, does not justify such a claim. This is wrong. I am glad we both agree that this is a false claim. If I boot 10 idling Mainframes on my laptop does not allow me to say "I have virtualized 10 Mainframes!! Look everyone!!!"

.

So let me ask you: when IBM says exactly the same thing, you dont object. Why?

IBM claims that one big Mainframe can virtualize 1.500 of the x86 servers. If you do some research, it turns out that all the x86 servers must idle at 1-2% (and they are old servers), and the Mainframe be loaded to 100%! Now, isn't this a false statement and FUD from IBM??? The Mainframes are not as powerful as IBM claim. That is pure FUD.

.

Likewise, IBM claims that one Mainframe can handle 400.000 logged in users. I am willing to bet money, that if I do some research, it turns out that all the users all idle.

.

Also, the legendary IBM Mainframe uptime. In one of scandinavian largest IT sites, a government manager was interviewed, he said something like "our mainframe crashed, that is weird. It never happens normally. Last time it happened, was 6 years ago".

Also, IBM is dead scared of the software emulator TurboHercules. An 8-socket Intel Nehalem-EX server gives 3.200 MIPS under software emulation with TurboHercules according to wikipedia. Software emulation is 5-10x slower than native code. Thus, the Intel server should actually give 16.000-32.000 MIPS. The biggest z196 Mainframe which has 20 cpus, gives 50.000 MIPS. Thus, you need 16 Intel Nehalem-EX cpus to outperform the z196 Mainframe.

The purpose of the TurboHercules software emulator, is that IBM customers can fail over to a x86 server when the Mainframe crashes. It is too expensive to buy another Mainframe. If the Mainframe never crashes, why is IBM threatening TurboHercules? There should be no market for TurboHercules, if the things IBM said were true.

And, a Mainframe has never been hacked? Pffft. That sounds like something IBM would claim.

@Sysprog Steve

This article is quite IBM biased? I have never seen strong Sun or Oracle biased article here, but Ive seen lots of IBM biased. For instance about the "super fast" POWER6 and POWER7. And of course, the IBM Mainframes.

.

However, regarding the POWER6: you need four POWER6 at 4.7GHz to match two Intel Nehalem (not -EX) at 2.93GHz in LINPACK.

You need six POWER6 (including p570) servers with 14 POWER6 cpus at 4.7GHz to match one Sun T5440 with 4 Niagara T2 cpus at 1.4GHz - when we talk about SIEBEL v8 benchmarks.

(You need 13 of the IBM CELL cpus at 3.2GHz to match one 1.4GHz T2 when we talk about String Pattern Matching). So, the POWER6 is not maybe the super fast cpu as claimed by IBM.

.

Regarding the POWER7. The Niagara T3 holds several world records. The Intel Westmere-EX is ~10% slower than POWER7. Maybe the POWER7 is not that super fast.

.

Regarding those "super fast" IBM Mainframes. As I have shown earlier, the IBM Mainframe cpus are quite slow. Any high end x86 cpu is faster. An 8-socket x86 server with Nehalem-EX gives as much as compute power as the biggest z10 Mainframe with 64 cpus.

Recently, IBM has released the "worlds fastest cpu" IBM Z196 running at 5.26GHz, 300MB cache (L1+L2+L3):

http://www-03.ibm.com/press/us/en/pressrelease/32414.wss

But, the z196 is only 50% faster than the slow z10 CPU, despite some crazy specs. Which means you need 12 Intel Nehalem-EX to match the biggest z196 IBM Mainframe cpu today.

Now, here is my question: how can IBM claim that you can virtualize 1,000s of x86 servers on a big mainframe? Answer: All x86 idle at 1-2% and the Mainframe is fully loaded. Well, in that case, my laptop can virtualize 10 IBM Mainframes. I just start up ten instances of the Mainframe emulator "TurboHercules" and load the Mainframe OS and idle them. I can emulate a Mainframe on a laptop. Google on "turbohercules".

.

Despite all those slow IBM gear, there are lots of strong IBM biased articles here claiming the gear is invincible and "never hacked" . How come? Why not such articles about Oracle? Can we see such Oracle articles too, maybe? A request from the readers here!

IBM to snuff last Cell blade server

@bazza

Yes, CELL was certainly a programming challenge, even for IBM.

For instance, IBM researchers did heavy optimizing (loop unrolling, asm, etc) in String Pattern Matching benchmarks, and still you need 13 of the CELL@3.2GHz to match one Niagara T2+2@1.4GHz. The Niagara did a plain implementation of the Aho-Corasick algorithm in pure C with no optimizations.

So the performance of the CELL might be stellar. But in reality it was quite slow and even researchers had serious problems to extract any performance.

Japan takes the Top 500 lead with K super

@A.C

Sure it runs 2048 cores, just as the ALTIX SGI server does. But it just a bunch of PCs on a switch.

Let me ask you: have you thought about this?

IBMs biggest Unix server P795 has 32 cpus

IBMs biggest Mainframe z196 has 24 cpus

Oracles biggest Unix server M9000 has 64 cpus

HPs biggest Unix server (Integrity?) has 64 cpus (I think)

And they fiercly fight for benchmarks. IBM was so proud over their P595 TPC-C benchmark world record. Why can not IBM simply put in 64 cpus in the P795? Why had IBM to rewrite the old and mature Enterprise AIX that ran on big Unix servers for decades - when P795 was to be released? The P795 has 256 cores, and that was too much for AIX to handle. The earlier P595 Unix server had 128 cores which was manageable by AIX. Why dont IBM put in 64 cpus? Or even 128 cpus? Are there some difficulties when you dont do clusters?

Why does Linux stutters on SMP servers with 32-48 cores?

@trstooge

Yes, but Todd Rundgren is correct.

These super computers are basically a large cluster on a fast switch. You just add a new PC to the network, and voila, you have increased performance. So, it has nothing to do with scalability when we talk about one large SMP computer, such as a IBM POWER795 with as many as 32 POWER7 cpus. Or Oracle Solaris M9000 server with as many as 64 cpus.

When we talk about one single large SMP computer, Linux is never run on them because Linux scales bad vertically. Linux scales to ~32 cores or so, on one large server.

Linux scales excellent in a large cluster with lots of PCs (good at horizontal scaling), but scales extremely bad on one single large server (vertical scaling). Linux merits is on a large cluster. Google runs a large cluster with Linux servers. There exist no Linux server with as many as 32 cpus. But there exists large super computers which are basically a cluster, for instance the SGI Altix with 1024 cores - which is just a bunch of blade PCs on a fast network.

@Todd Rundgren

Do you have some links on your claim? I would like to read more, please.

Time to say goodbye to Risc / Itanium Unix?

@Jesper Frimann

You are just evading the issue.

I am claiming that POWER gets cheaper and cheaper, and Intel gets faster and faster. Then I gave some numbers. I said POWER7 is 10% faster than Intel. It might be 13% or 15%, so what? The trend is clear: POWER6 was several times faster than Intel before (on some benchmarks) and costed 5-10x more. POWER7 is only slightly faster, and much cheaper.

Instead of disputing the exact numbers I suggest you discuss the issue instead: POWER is getting cheaper and cheaper, and Intel is getting faster and faster. This is true, it is a fact. Intel cpus are getting more performance at a quicker pace than IBM POWER is getting performance. The Intel slope is higher which means the lines will cross: Intel will catch up on POWER.

Also, IBM has said officially that AIX will be killed. This is also true. It is a fact.

.

Now, here comes the guess part (these are not facts, but opinions):

I predict that when Intel is faster, or fast as POWER cpus, then POWER need to lower the price even more. And finally when the price is too low so that IBM does not earn much money anymore, POWER will be killed off. Just like IBM killed of CELL cpu. And that is the time when AIX finally will be killed by IBM. AIX is soon no more high margin business.

I am basing this prediction on IBMs official statements and I am extrapolating the clear trend (POWER getting cheaper, and Intel getting faster). In other words, I do have some substance in this prediction: it is not evil rumours without ground (not FUD). I am not FUDing.

@Jesper Frimann

The question is not the numbers, never mind them. It does not matter if it is 10% or 13%.

The question is: POWER is getting cheaper and cheaper. And Intel is getting faster and faster. Soon, POWER will be very cheap, to be able to compete with Intel, and that is when POWER will stop being high margin. And why would IBM do low margin business, when POWER costs billions in R&D? Why should IBM not kill a low margin business when you get a off the shelf Intel CPU at a lower price, and higher performance?

@Beachrider

I post links to IBM senior executives. Would you prefer me to not post links at all, so you could dismiss my posts as pure FUD and rumours? Well, IBM executives has said what I claim, I am only quoting them.

POWER6 was several times faster than x86 but costed 5-10x more.

POWER7 is 10% faster than Intel Westmere-EX and costs 3x more.

What is your predictions about future POWER cpus? What does the trend tell you? That IBM will continue to lower the price of POWER? That POWER is soon not a high margin business anymore? And you know what IBM does with low margin business? IBM kills it. As any sane capitalist company with greedy share holders would do. IBM is an american company. IBM gets rid of low margin business, when IBM spends billions on R&D and can not get the expends back. x86 is the future. (Even though x86 is a buggy piece of shit, it is cheaper and soon faster)

@Splitbrain

You should calm down, or you will get a heart attack.

But the trend is clear, POWER gets cheaper and cheaper. And x86 becomes faster and faster, and soon they will cross. Then why should anyone buy expensive POWER gear?

@A.C

"....you cannot quote a misquote from 2003 as fact..."

Am I "misquoting" the official statements from IBM? How am I misquoting them? I am trying to explain to Jesper Frimann, that according to IBM, AIX will be killed off and replaced by Linux. How is this a misquote? Please cite me, and the article I linked to - and if you can point out any errors in my understanding of the article I will stop say false things. If you can not point out my misunderstandings, then maybe it is you who misunderstood the official statements from IBM?

.

.

"...What has happened since 2003? Well AIX was 3rd in the market and has now not only passed Solaris and HP but is 50% of the Unix market. I think a lot has changed since Power4 and AIX 4.3...."

So? It seems that you dont really understand me. I say that long ago, POWER6 was several times faster than x86 and it costed 5-10x more. Today, POWER7 is ~10% faster than Intel Westmere-EX but costs 3x more. Future POWER cpus will cost 1-2x more, but ideally will be cheaper than x86 or they will not sell.

Yes, a lot has changed since POWER4. The trend is clear: AIX/POWER is not a high margin business anymore. IBM had to lower the price of their best CPU yet (the mighty POWER7) because IBM is afraid of the competition from x86 and Niagara. This means something.

I predict that IBM's fear will increase, and future POWER cpus will be cheaper and cheaper. There will be a time in the future when IBM POWER can not compete with x86. x86 will catch up on POWER. You know that. I know that. This is true. The trend is clear, it will happen. Some time after, IBM will kill POWER, because IBM can not afford to loose money on expensive and slow POWER cpus, with a diminishing market share. That is when AIX will be discontinued.

Do you really think that IBM will release a future POWER cpu that is many times faster than x86, so IBM can continue to charge 5-10x more? No. Those days are gone. x86 will be faster. And AIX does not run on x86. So AIX on POWER will be no more. This is true. Even IBM executives see this trend.

Next year, the 22nm Ivy Bridge from Intel will arrive. It will be 40% faster (according to Intel) than today's Sandybridge. Thus, it will be faster than POWER7. IBM must make sure that POWER8 is much faster than x86, there will be no point in spending billions on R&D on a slower POWER8 cpu.

So, AIX and POWER is soon obsolete. Itanium is obsolete (this is a pity). Better start to learn Linux on x86. The thing is, the managers dont understand that POWER and Itanium has much better RAS than x86, and that x86 is a piece of buggy shit. They just look at the cheapest performance / money, and x86 is the future. The managers will only consider OS that runs on x86.

@Jesper Frimann

"...Sure it's sad that in the server marked is looking like AIX and Linux are going to be the only members from the UNIX'es left in the long run..."

Maybe you missed it, but according to IBM executives, AIX will be killed off and replaced by Linux:

http://news.cnet.com/2100-1001-982512.html

Thus, it seems that IBM plans to let only Linux to survive. No more AIX. When AIX will be discontinued, most probably IBM will continue to support AIX for another 10 years. So dont worry, you will still have work.

Pity that IBM will not port AIX to x86, though. When x86 outperforms POWER, IBM must port AIX to x86 or AIX will surely die. And preferably, IBM should open source AIX to ensure it to survive by the community when IBM kills it. And POWER will also die, because no one will buy POWER to run Linux on it, when they can run Linux on x86 faster and cheaper:

http://www.anandtech.com/show/4285/westmereex-intels-flagship-benchmarked

POWER is today only ~10% faster than x86, but costs 3x or more.

.

Regarding Solaris, there are no outspoken plans to kill it. Actually, Larry Ellison is planning increasing investments in Solaris, far more than Sun ever invested:

http://www.informationweek.com/news/global-cio/interviews/222500081

And Solaris source code is opened, so it can not be killed by Oracle.

.

Regarding HP-UX, I have not heard any plans to kill it. Of AIX, Solaris, HP-UX and Linux, there are only official, outspoken plans to kill AIX. It seems that your conclusion is not in alignment with IBM senior executives.

AMD promises 10 teraflop notebooks by 2020

@Jim O'Reilly

"...And what do you do with 10 TeraFlops? Email?

Seriously, the amount of compute power is increasing so rapidly, that the mundane can be done with waste cycles. So, unless these super-books are only going to be used by 10 PhDs at the National labs, we need to cover some other bases..."

.

Computers are becoming faster, which allows us to use more complex software. Compare MS-DOS and the best word processor, vs Win7 + Word2010. The difference is vast. In the future, there will more than 1080p, and 3D GUI, and you can talk to the computer, etc etc which draws computer power. The faster the computer, the more complex software and the easier the life will be for us.

It almost sounds like "who need 640KB?".

HP threatens Oracle with legal action for jumping Itanic

@A.C

It is not meaningless. It just tells that Oracle is attacking Itanium by making it more expensive. And by coincidence, it turns out that Oracle has stopped development for Itanium - which really proves that it is an Oracle attack on all fronts.

Your statement would be true, if Oracle still continued to support Itanium. Because Oracle would reason something like this "we can not compete with Intel performance wise so we have to make them more expensive". But no, Oracle is attacking on fronts - it has nothing to do with performance problems.

BTW, Oracle Niagara cpus happens to have several world performance records today. I would not call it "slow".

@Ioan

I do not always agree with Splitbrain, but in this matter he is correct. The database is the heart of a company, all data is stored there. Without the data, a company is nothing. And you can not just do a heart surgery "with a flick of hand" as you claim. If you really know how to migrate from Oracle to IBM DB2 in a quick and easy way, I suggest you write a book or start a consulting firm - you will be very rich.

Your post reminds me of the usual IBM FUD "I work at a large bank / stock exchange / biggest company in the world, and we love SPARC, but recently we tried IBM POWER and wow they are cheap and fast. No matter how much we love SPARC we must now migrate to SPARC. Too bad. We really really love SPARC but POWER is just so much better. I advice everyone to do the same. Migrate to POWER, or you will go bankrupt"

It is just a another variant of the normal IBM FUD:

"It is easy to migrate from Oracle to IBM DB2, anyone can do it. We love Oracle, but it turns out that IBM DB2 is better and cheaper, and easy to migrate to". Highly dubios.

Server sales grow thanks to big boy boxes

@Jesper Frimann

I dont understand what you mean. Oracle say something like this "On october 14th we will show you how fast we are and give you the exact numbers. Just wait some more"

Where is the FUD in that? Oracle revealed the exact numbers and performed better than IBM - just as Oracle promised. Oracle did not break any promises. Oracle did not lie. Oracle did not say "we will show you a non clustered result that is faster than IBM" - no, Oracle said nothing about clusters vs non clusters. Oracle has not lied about that - because Oracle said nothing about that.

The Oracle poster would be FUD if it turned out that IBM was faster, because then Oracle would have lied. That Oracle poster is not FUD. You dont know what FUD is. FUD is basically, when you lie. If you have negative things to say, it might be a valid point and important to know. Negative critisicm is not FUD.

Negative lies is FUD, just read the article again; IBM said negative things which where not true - THAT is FUD. If IBM had said negative things that were true - that would be valid criticism and that would be important for the customers to know. Then IBM would be right in telling their customers no to use non IBM gear. But it was not valid criticism, it was lies. Ergo, FUD.

@Splitbrain

I do not try to hide or conceal that I am an Oracle supporter. But I dont like all the FUD from IBMers. I try to counterbalance that. I have been accused of providing too many links and articles - well that is a good thing from me. If I did not provide links, I would be accussed of FUDing and making up things. But I do not, which is proved by all my links.

IBM was the first company to systematically employ FUD:

http://en.wikipedia.org/wiki/Fear,_uncertainty_and_doubt#Definition

It is natural that the IBM supporters FUD a lot, with such a master...

If you disagree with my posts, how about you try to argue with facts and links, that I am wrong? I have noted many times, in many debates, that I provide lots of links and proofs of my view point - whereas the opponent never provide any links at all, they just yell at me. Sometimes they provide links, which is good. But not always.

.

Regarding if Sun/Oracle FUD and lie. Someone said that Sun canceled Rock, canceled this and that. Well, that is not a lie. IBM releases a product, and claim it is 100x faster than it actually is. THAT is a lie. IBM claims that their mediocre hardware can virtualize 100s of other servers. If you read the foot note, the other servers are typically 1GHz cpu with 256MB RAM or something like that - and they all idle. Now THAT is a lie. Very typical from IBM.

@Beachrider

IBM Mainframes are not supercomputers? So you, too, admit they are not suitable for number crunching, and in fact, they are very weak in that area? Any x86 cluster would outperform a Mainframe by far. So it seems we all agree that cpu performance is not the strength of Mainframes. (How in earth can a low performant z10 Mainframe virtualize 1.500 x86 servers? I dont get it, imagine if 10 of the x86 servers start to do some heavy cpu work)

.

Other touted strengths of mainframes, are the reliability and I/O.

Regarding the reliability, I am doubtful. For instance, in Scandinavias largest IT site, there was an article of a big company whose Mainframe crashed, and the managers said "unfortunately our IBM mainframe crashed, it never crash usually. Last time it crashed was six years ago".

http://www.idg.se/2.1085/1.363240/stordator-i-oslo-bakom-sjs-stora-problem

And the purpose of the software emulator TurboHercules is that companies that can not afford another IBM Mainframe, can use TurboHercules for failover when the Mainframe crashes. The thing is, the x86 servers using software emulation (which is 5-10x slower) still rival the IBM Mainframes in terms of cpu performance. 8-socket Intel Nehalem-EX gives 3.200 MIPS - which is a decent sized Mainframe. For a fraction of the price. You can emulate a Mainframe on your laptop.

Now, if Mainframes never crash, why would IBM try to stop TurboHercules? There should be no market for TurboHercules. If the Mainframes are superfast, then software emulation could never give decent Mainframe perfomance - so IBM would not try to stop TurboHercules. But IBM is vigorously fighting TurboHercules. IBM is afraid of TurboHercules. Founder of Hercules describes what IBM does to his company:

http://herculesvsibm.wordpress.com/

.

Regarding the massive I/O. Sure, Mainframes have good I/O. They have like 296.000 I/O channels or so. One Mainframe can handle 500.000 logged in users. That is impressive.

But IBM claims that one z10 Mainframe can virtualize 1.500 x86 servers. In the foot note, IBM assumes they all idle. So... if a Mainframe can handle 500.000 users, maybe the vast majority idles? Maybe only 1% does actual work? I would not be surprised if IBM marketing lied about this too.

Regarding finance and algorithmic trading and superfast HFT. Mainframes are never used. They are good at batch work. High throughput. But crappy latency. In algo trading, etc, they typically all use x86 servers with Linux/Solaris.

.

To recap about the Mainframe virtues:

1. Superfast. No, they are not. They are slow - when we talk about cpu performance.

2. Reliable. Why do they crash every 5 year? Why do people want failover software emulation?

3. Good I/O. Maybe 95% of all users idle?

4. Backwards compatibility. I know too little about this. But I suspect this is true. Enterprise tend to focus on backward compatbility. And Mainframes are Enterprise.

5. High price. Sure. This is true. But any POWER7 machine is faster than a IBM Mainframe. Let alone decent x86 server.

Correction

"...IBM will support AIX and POWER for another 10 years - even though development has stopped..."

Of course I meant "even IF development has stopped". I do not mean that development of AIX and POWER has stopped. My entire post talk about the future, "when IBM kills off AIX and POWER". It is pretty clear that AIX and POWER has not been killed, if you read my post. I talk about the future to come.

When IBM finally do what they say they will do and kill their AIX and POWER, I will write it in every post. In big letters, so no one miss that. But not yet, it is in the future.

Allison Park

"...Oracle and Fujitsu are really good at making big announcements without content...."

At least they have substance, which IBM does not have.

When the Mainframe cpu is much slower than a Nehalem-EX, how can IBM claim that a 20cpu Mainframe can consolidate 1500 x86 servers? That is just not correct. Some would say it is a lie.

Also, IBM claims that their new Mainframe Z196 cpu is the "worlds fastest". It has impressive specs; 5.26 GHz and almost 300MB of cache (L1+L2+L3+L4). But actually, it is much slower than a cheap Intel Nehalem-EX - as I have proved earlier. So how can it be "the worlds fastest cpu" when any modern x86 cpu is faster, and cheaper? Is this also a lie?

http://www-03.ibm.com/press/us/en/pressrelease/32414.wss

Can you quote lies from Oracle? I want to see them? I have quoted lies from IBM, you do the same now.

.

Regarding the new Intel 32nm server cpu Westmere-EX. It seems that the POWER7 could be something like 15% faster. But the POWER7 costs 3x as much, or more.

http://www.anandtech.com/show/4285/westmereex-intels-flagship-benchmarked

IBM has made hefty price cuts for the POWER7, beacuse IBM is afraid of the competition. IBM knows POWER7 will be caught up on performance, very soon. IBM cut the price, because they know there is no incitament to buy slower but 10x expensive IBM gear like in the old days. Now IBM only costs 3x as much (entry level).

In the future, IBM will be forced to cut prices even more to compete with others. Soon POWER will costs as much as the competitors. This will happen - we see the trend, POWER is becoming cheaper. Because it is not faster nor better than x86. Then why should IBM continue development of POWER? It will be easier to buy an x86 which will just as powerful as POWER, at the same price (which will be very low). No more high margins for IBM POWER. Margins will be very low, and compete with the huge x86 market. When that happens, IBM will just drop POWER.

It seems that my prediction comes true: future x86 will catch up on IBM POWER cpus in terms of performance, because x86 develops much faster than POWER. When x86 is faster (very soon) then IBM will officially kill off AIX - just as the IBM executives officially said:

http://news.cnet.com/2100-1001-982512.html

IBM AIX development has slowed down, for a reason: IBM is preparing for the killing of AIX. Of course, POWER will also go down the pipe.

But IBM customers dont need to worry, even if Itanium is killed off today, HP has promised to support it for another 10 years or so. IBM will support AIX and POWER for another 10 years - even though development has stopped.

.

Regarding Solaris, it already runs on x86 and does it well. If SPARC is killed off, Oracle will just continue to sell x86 servers with Solaris - just as today. But, Niagara is many times faster in it's niche, so x86 can not compete in the niche. Niagara will be there, as it delivers (in the niche). IBM CELL did not deliver, it was killed off. POWER to follow in a near future. IBM AIX is not ported to x86, it does not run there. And IBM has said it will be killed off.

Server biz bouncing back to boom times

@A.C

You IBM supporters really feel uncomfortable when I remind you what IBM executives has said about AIX, dont you? You are not too happy about that, I should shut up, right? How can you make shut up? Ask me not to post that official statement from IBM? No, it would not work. How about you IBMers mock me? Could that work? No, you are dealing with Kebabbert, and he only accepts facts and proofs.

If you want me to stop telling everyone that AIX will be killed, then it is best if you disprove it with facts and links. Then I stop. Mocking me will not work.

And by the way, have you heard that IBM has officially said that AIX will be killed? That means that POWER also will be killed. Just like IBM CELL cpu was killed - because of abysmal performance.

@Jesper Frimann

"...But why don't we just be a little happy that the UNIX marked has had a terrific quarter, be it AIX, HPUX or Solaris. That should be something that everyone could agree on is a good thing. Well perhaps all but kebbabbertie...."

I dont agree with you. I remember when next gen Itanium posts where FUDed by IBMers. "Itanium is slow and buggy, migrate now to POWER7 before it is too late", "we have migrated off Itanium to POWER7 and everything is much faster and stable" - and someone asked "how could you, the POWER7 servers are not for sale yet!".

With IBMers track record here regarding HP-UX, I doubt IBMers would be happy if HP-UX grew. Or if Solaris grew. Only AIX growth would be accepted - which would be taken as a reason to further trash talk HP-UX and Solaris.

If you IBMers stop FUD, then I stop post that "AIX will be killed, said IBM executives"

@A.C

So IBM does not have to have a public roadmap, but the other Unix vendors must have? Why? Oh, you are refering to the IBM executives that officially has said that AIX will be killed off? Ok, now I understand. That makes sense.

http://www.zdnet.co.uk/news/application-development/2003/01/29/ibm-linux-will-replace-aix-2129537/

.

POWER6 was 10x more expensive than x86 gear. POWER6 was considerably faster than x86 cpus.

.

POWER7 is only 3x more expensive than x86 gear. POWER7 is 10% faster than the new Intel Westmere-EX, according to anandtech benches:

http://www.anandtech.com/show/4285/westmereex-intels-flagship-benchmarked

.

The next gen IvyBridge 22nm Intel cpus will be considerably faster than today. Some say 30% faster. IvyBridge will be faster than POWER7, at a fraction of the price.

Future Intel cpus are catching up on POWER very fast, and is today already able to compete with POWER. POWER8 might be the first POWER cpu to be slower than Intel x86 cpus. This means IBM have to continue to lower the price on POWER8 servers. The margin is gone. Why should IBM spend loads of money to develop a slow POWER cpu? Why not use faster and cheaper x86 cpus?

That will be point when AIX will be replaced with Linux. This will happen. IBM has said that officially. I dont see the point of developing POWER, and believe it will be killed too. Just as the "superior" IBM CELL has been killed (because of bad performance) - POWER will be killed off too.

.

Ergo, the reason IBM has no public roadmap is because AIX will be killed off, and probably POWER too.

But dont worry, Allison Park, you will have work in the future. When IBM kills off AIX and also probably POWER, IBM will continue to support them for at least 10 years more. So you will continue to have work too. After that, you better start to learn Linux if you want further employment.

On the other hand, Larry is committed to Solaris and SPARC. On it's niche, Niagara is much faster than x86. x86 will be faster general purpose cpu, yes. But Niagara is a niche cpu and will thrive in the niche. For general purpose, Solaris runs fine on x86. AIX does not exist on x86, and it will not be ported to x86, because IBM will stop development of AIX in the future. No point in spending money on porting AIX to x86 when AIX will be killed.

IBM guns down Neon's mainframe accelerator in Texas

Dell

should not allow anyone to plugin a 3rd party graphics card. You should only be allowed to use DELL hardware, or be sued. That would be good business for DELL.

As I understood it, you lease the slow IBM Mainframes, so you dont own them? If you bought them, then it's yours and you can do whatever you want with it.

I hope this teaches a lesson for new companies to not rely on Mainframes. There are no new companies basing their business on Mainframes. Only old companies upgrading.

Sinking Sun scuppers Oracle server figures

@Beachrider

I do not agree with much you say, some things are just plain wrong.

"...IBM is prepared with higher GHz CPUs, higher thread-count processors (96/socket by Fall) and fancy I/O for Power 7+. ..."

Dont you know that GHz does not necessarily imply good performance? Do you remember Intel Pentium4 which had high GHz, and very low performance?

Or, do you happen to know about the new IBM z196 Mainframe cpu "the worlds fastest cpu" at 5.26GHz with 300MB cache? It has lower performance than a Intel Nehalem-EX and costs much more.

Have you heard about the slow POWER6 at 5GHz? You needed six POWER6 servers with 14 POWER6 cpus at very high clock, to match one Sun T5440 with four 1.4GHz Niagara T2+ when we talk about Siebel 8.0 benchmarks.

Or, the IBM CELL. You need thirteen CELL cpus at 3.2GHz to match one Niagara T2+ at 1.4GHz in string pattern matching. That is, 42GHz worth of aggregate GHz to match 1.4GHz. That is a factor of 29.7. You need ~30 times more GHz if you use IBM cpu CELL to match one Niagara T2+. that is quite bad.

So you are wrong when you imply that high GHz equals good performance. IBM gear disproves your claim.

.

Regarding IBM having higher thread count processors (96/socket) - what do you mean? I mean, Niagara T3 has 128 threads per cpu. To me it seems that 128 threads are more than 96 threads. Maybe you dont know about Oracle cpus? Niagara T3 has several world records today, beating POWER7 in some benches.

.

.

"....[IBM] will NOT be dropping the price dramatically, though. IBM's investors were clearly surprised by the price-drop in early 2010 and have exerted pressure to not-have IBM attempt to make Power 7 a dollar-to-dollar competitor to Intel's commodity business. They want Power to sell at a substantial premium over x64...."

What are you talking about? POWER6 servers costed like 5-10x more than x86 servers. Now the POWER7 costs 3x more than x86 servers. Today the Intel Westmere-EX 32nm is something ~10% slower than POWER7:

http://www.anandtech.com/show/4285/westmereex-intels-flagship-benchmarked

The 22nm version will be 40% faster according to Intel. In that case, it will be faster than POWER7. IBM has no choice but keep lowering their price. Why would anyone buy a slower and 3x more expensive POWER7 server than a cheap faster x86 server?

In a couple of years, Intel and AMD 20core cpus, will easily be faster than everything IBM has. That is when IBM will see their POWER sales dwindle, and they start to lower the price. Finally, x86 will be very much faster and cheaper - that is the point where IBM will kill POWER and AIX. As IBM executives officially said: they will kill off AIX.

http://news.cnet.com/2100-1001-982512.html

Then there is no reason to continue development of POWER. It will be killed off too. IBM must continue lower prices in the future. Gone are the high margin AIX systems. Left are the slow IBM Mainframes which has a high margin.

As I said, I dont agree with you much. Much of what you say is errorneous.

IBM preps Power7+ server chip rev

@A.C

So what are you trying to say? What is your point? That the benchmark is made up and never done? It is a lie? Well, it is easy for you to implement the same well known algorithm and run it and see. I never doubt that the numbers were actually achieved, because that would mark the source as a liar, and no one wants that. It is too easy to repeat the same benchmark and disprove the benchmark.

@A.C

Well, I work in finance. A large world famous company. And I KNOW we dont use floating numbers. We use integers and avoid rounding. I suggest people to start to use integers instead of decimal floating numbers. To round is amateurish at best, and wrong.

@Jesper Frimann

This is my discipline. I work here. You are wrong.

@bazza

I dont agree with you much, bazza.

"...It's because for banking / financial applications standard double precision FPUs on Intel/AMD chips are not accurate enough, so you have to do the maths a different way..."

So, that is the reason POWER has a decimal math and standard FPU? Well, you are wrong. Financial and banking applications never do any rounding, they never use an FPU. Everything is computed as integers, and the number of decimals are book keeped separately. So, when you do a computation, you specify the amount of money, and you specify the nr of decimals. This way you dont never do any rounding. If IBM POWER are doing rounding, they are amateurs having no clue about finance nor banking.

.

Regarding why IBM stopped manufacturing the "superior" CELL cpu, you are wrong again. It is not because people could not appreciate nor program the CELL tech, as you claim. That is just IBM marketing talk. The reason CELL stopped, was because it didnt live up to expectations. It was bad, quite simply.

For instance, in String Pattern Matching, a heavy optimized CELL version (loop unrolling, assembler, etc) done by IBM researchers, bit the dust when compared to a SPARC Niagara T2+. The T2+ running at 1.4GHz is 13x times faster than a 3.2 GHz CELL. The T2+ just used a plain C version of the algorithm, copied from an algorithm book with no optimizations at all. It is pretty bad when a 3.2GHz using heavy optimizations by researchers can not beat a 1.4GHz Niagara, dont you think? You need more than 10 CELL cpus totaling 32GHz of aggregate Hz, to match one Niagara which has 1.4GHz. The CELL construction is very ineffective. One Hz from Niagara is as fast as 23 Hz from CELL. The ratio is 1:23 - which is not very effective at all.

http://blogs.oracle.com/bmseer/entry/extremely_fast_pattern_matching_on

It is just IBM that tells everybody that CELL is superior, but people dont have the wits to appreciate nor program it. The truth is, the CELL sucked quite bad. Otherwise it would not be discontinued.

Unisys revs up big ClearPath mainframes

@David Beck

"...Also, how the heck do Xeon processors provide either of these environments? Straight emulation/simulation? The entry machines must spend most of their time decoding and shifting bits about...."

If these Mainframes are similar to IBM Mainframes, then it should be no problem to emulate them The TurboHercules emulator for x86, can give 3.200 MIPS on a 8-socket Nehalem box. The biggest IBM mainframe is 10x-15x faster. That is not extremely good a result from IBM, considering that IBM Mainframes run native code, and the x86 runs software emulation which is 5-10x slower. Also, the IBM Mainframe has 20 cpus vs the 8-socket Nehalem box.

In short, the x86 should be able to emulate Mainframes very fast. If you ported the software to x86, you would get 5-10x higher performance.

HP forges 32-socket Itanium iron

@Jesper Frimann

I am just trying to say to all the Linux fanboys that the SGI Altix machine is basically a big cluster and it is not comparable to a big SMP machine such as P795 or the M9000. But the Linux fanboys dont believe me - they say that Linux scales better than AIX and Solaris together. Which is wrong. Scaling a cluster is easy.

Regarding the M9000 vs P795 - everyone agrees that the POWER7 is superior to the CPU in M9000. That is not the question. The question is if the Linux fanboys can claim that Linux SGI Altix server with 256(?) cpus is a SMP machine. It is not.

.

Regarding the Itanium. Here are some interesting links. Intel is relocating the Itanium engineers to Xeon.

http://www.xbitlabs.com/news/cpu/display/20110503125538_Intel_HP_Unsure_About_the_Future_of_Itanium_Kittson_May_Be_Last_Survivor.html

http://www.xbitlabs.com/news/cpu/display/20110417112218_Intel_Relocates_Itanium_Engineers_to_Xeon_Projects_Sources.html

Intel claims that Xeon has better RAS than Itanium:

http://www.xbitlabs.com/news/cpu/display/20110413162720_Intel_Xeon_Now_Has_Equal_or_Better_Reliability_and_Performance_than_Itanium.html

I really doubt that, as x86 is too buggy. I really believe Intel wanted to learn RAS from HP, and then kill off Itanium. Which is quite dumb, because x86 is too buggy and will never has as good RAS as Itanium.