Feeds

back to article IBM gloats over HP, Oracle takeouts

Every quarter, a few weeks after IBM puts out its financial figures, it makes the rounds with the trade press to gloat over how many takeout deals it did, removing gear from its rivals and usually replacing it with System x or Power Systems iron but sometimes with a mainframe. When Mark Loughridge, IBM's chief financial officer …

COMMENTS

This topic is closed for new posts.
Anonymous Coward

IBM just needed to sit back and wait!

From a customers perspective: It is very clear why so many companies moved over to IBM. During the reign of Hurd at HP, they closed regional offices, retrenched key technical staff and cut R&D. Simply put, HP no longer had the capacity or capability to support their customers anymore. A simple case of penny wise pound foolish.

As for Sun Microsystems – Customers certainly know all about Oracle’s high margins and questionable value proposition. Its simply a matter of get out now while you still can.

IBM just had to sit back and wait..... customers no longer had the option but to move!

2
0

Pure FUD

"As for Sun Microsystems – Customers certainly know all about Oracle’s high margins and questionable value proposition."

Oracle have high margins, and at the same time talk about IBM? That must be the weirdest thing I have heard. IBMs Mainframes takes 10% of all money of the entire server market, and number of Mainframe shipments is down to 0.01% of the entire server market. Now THAT is what I call a fat margin. 0.00% of all servers, take 10% of all money. This, I read here on theregister.co.uk.

.

Also talking about "questionable value proposition" - you needed 6 (six) IBM Power P570 servers to match ONE Sun T5440 server in SIEBEL v8 benchmarks. And ONE P570 server costed $413.000 USD, whereas one T5440 costed $76.000 USD. You needed IBM servers worth of 6 x 413.000 = 2.5 million USD to match one Sun server. Now that is what I call questionable value proposition from IBM.

Not to mention those uber slow Mainframe CPUs. One of the newest z196 Mainframe CPUs, which IBM calls "worlds fastest CPU" at 5.2GHz and 300MB cache (L1 + L2 + L3 + L4) is really dog slow. You need 5-10 of the z196 cpus to match one modern x86 cpu. How can IBM say that the z196 is the "worlds fastest cpu"?

Oh wait, of course, now I understand. IBM and IBM supporters have a weird logic, that is the reason!

Earlier, when Oracle had the TPC-C world record, IBM said that because IBM has faster cores, the TPC-C world record still belonged to IBM. If you looked at no 1 on the TPC-C list, you saw Oracle. But still IBM said that they had the record, because they "used faster cores".

Or, the POWER6. You need four POWER6 cpus to match two Intel Nehalem - and guess which cpu is the fastest? POWER6 OF COURSE! Because "the POWER6 core is faster" - yes that is true! They say that! Weirdos.

Against such a logic, how can you explain they are wrong?

0
8
Headmaster

Talking about FUD

"Oracle have high margins..."

Yes they have very high margins.. on Software. On the hardware I don't think they are high enough. I don't think that Oracle is actually making money on their server sales. And besides that direct hardware revenues it only accounts for 4% of Oracles revenue.

http://www.oracle.com/us/corporate/investor-relations/financials/q3fy10-detailed-financials-080347.pdf

On page 3:

Oracle made 458 Million USD on direct Hardware revenue + Support. And lets give them 10% of the services revenue that it 10% of 2,797 for a total of... 738 Million USD.

But then comes the expenses related to making that revenue.. and that is when the bad story begins. Again if we look at the above link there are direct expenses and % of other expenses (Lets be real nice and use 4% it's in reality much higher and 10% on services)

Expenses:

Hardware systems products 206 MUSD

Hardware systems support 116 MUSD

Sales and marketing 133 MUSD

Research and development 88 MUSD

General and administrative 25 MUSD

Amortization of intangible assets 55 MUSD

Services (10%) 243 MUSD

-------------------------------------------------------------------------------------

Expenses 866 MUSD

Revenue 738 MUSD

-------------------------------------------------------------------------------------

Loss 128 MUSD

===========

And then comes the restructuring cost and acquisition costs relating to the buy of SUN.

Which for Oracle as a total was a total of 517 MUSD.

Lets just say that 25% of that was related to the SUN's hw business that would then put the Oracle loss on the old SUN HW business at 257 MUSD.

And IRL the actual number is much bigger, cause it's a much more labour intensive job selling hardware than software or rather in Oracle's case software maintenance which is a stunning 56% of all Oracle revenue, so the 4% I used is most likely much higher, at least 8% which would put the loss of over the 500MUSD mark.

Now such a big loss is not something that Larry will put up with in the long run.

This reminds me when I was a Freelance consultant 13 years ago, and advised my client against making Digital Servers the strategic platform for the next 5 years, cause the company was in trouble. I wasn't popular but I was right.

This is most likely the story that IBM sales is telling consulting companies like Accenture,McKinsey, KPMG, CapG etc etc. And that will hurt Oracle, if these 'independent' advices start to put the thumb down on Oracle HW.

"lso talking about "questionable value proposition" - you needed 6 (six) IBM Power P570 servers to match ONE Sun T5440 server in SIEBEL v8 benchmarks. "

Again you keep repeating that story again and again.. it's 2/3 filled up POWER 570 and a 1/8 th filled up POWER6 p570 versus one T5440. Hence your claims are ... false.. and stink of FUD.

And rather than just quoting the now gone BMSEER you should try to look at the facts.

http://www.oracle.com/apps_benchmark/doc/sun-siebel-8-14000-pspp-on-solaris-benchmark-white-paper.pdf

and

http://www.oracle.com/apps_benchmark/doc/IBM_Siebel8_7000_PSPP_On_AIX_POWER6%20Final.pdf

Now the reason why the T5440 is on top here is the response times, which is 4 times faster for the majority of uses on POWER6/POWER5 solution. Which someone who actually worked with RL IT systems, should know.

"is really dog slow. You need 5-10 of the z196 cpus to match one modern x86 cpu."

Now that is just pure FUD, you have no what so ever facts to match that statement. And IMHO it is clearly untrue. And shows that you don't know what you are talking about.

"Earlier, when Oracle had the TPC-C world record, IBM said that because IBM has faster cores, the TPC-C world record still belonged to IBM. If you looked at no 1 on the TPC-C list, you saw Oracle. But still IBM said that they had the record, because they "used faster cores"."

BZZZZZ.. Wrong again.. Oracle had the 'clustered' TPC-C record, not the single Machine. Now they don't have the clustered record, it is also a POWER score.

"Or, the POWER6. You need four POWER6 cpus to...."

You do know that POWER7 have been shipping for a long time don't you ?

Or is the only thing you can do whine FUD, come on grow up learn to research your arguments and do some logical reasoning. It's pathetic.

4
0

@Jesper frimann

Jesper Frimann

It is typical of IBMers like you, to not do fair comparisons. Oracle has recently bought Sun, it takes time to get their hardware business going smooth again. What you say, proves nothing. When things have been stabilized we can discuss this again.

But of course, you claimed that when Niagara T2 got a Hz upgrade from 1.4GHz to 1.6GHz - you claimed it was the next gen cpu and it should be compared against POWER7. If you can claim that, why can not you not talk about a company turmoil where things are just beginnning to sort out? Of course you do! Oracle will gain ground later, when they have fully integrated Sun.

It is like if I compare a beta version (not finished nor polished) of IBMs stuff, to a fully mature product to Oracle. Of course IBM's product would loose benchmarks, but that is an unfair comparison and proves nothing, it is not relevant to the discussion. You have to compare generation to generation - to see which tech is best and fastest. Not compare about old tech to new tech - that is not fair (unless it is IBM or IBMers that do the comparison)

If I would have done such an unfair comparison, you would have screamed loud. But when you do such a comparison, it is fair and good. A bit biased, yes?

Also, IBM power systems revenue went down 13%.

.

"...Again you keep repeating that story again and again.. it's 2/3 filled up POWER 570 and a 1/8 th filled up POWER6 p570 versus one T5440. Hence your claims are ... false.. and stink of FUD..."

Ok, I did not know that there is a difference between IBM p570 and IBM POWER 570 - I thought it was the same server. So, I should instead have said: "you need six p570 and POWER 570 servers to match ONE Sun T5440 server on official SIEBEL benchmarks v8.0" - then my statement would have been correct. So, I dont see how I FUD? Do you deny that you need six IBM servers to match one Niagara server?

You posted two links:

In IBM's white paper on Siebel v8, they used three IBM POWER servers: p570 and POWER 570. IBM had in total twelve(12!!!) POWER6 cpus at 4.7GHz and another two POWER6 cpus at 1.9GHz. In total 14 POWER6 cpus, and IBM got only 7.000 points. This only proves that POWER6 is a piece of sh*t cpu - and you FUDed a lot about how fast it was. Lesson learned: Never trust a Jesper Frimann. Especially not when he claims that even though you need four POWER6 to match two Intel Nehalem - the POWER6 is faster. That is not even FUD from Jesper, that is an obvious lie. You LIE, Jesper!

In Sun's white paper on Siebel v8, they used one T5440 with four(4) Niagara at 1.4GHz and got 14.000 points.

So if I claim that you need six IBM servers to get 14.000 points - do I FUD and lie (like you do all the time)? No I dont. I can always back my claims up with white papers and benchmarks, which you always reject with excuses, see here:

Then you start your excuses, and start talking about IBM having faster response times, and that is the reason IBM gets only 7.000 points.

What a bunch of horse sh*t. As of now IBM has the TPC-C record which I dont deny nor find sille excuses for. I accept your benchmarks and white papers. But you never accept mine. Just like in this very post - which you prove again here.

I talk about white papers (which you showed here) from IBM and Sun, and you immediately deny them. Call them FUD and false statements, and find excuses: "IBM has faster response times" so the benchmarks are not valid, bla bla bla. It sounds like "IBM still has the TPC-C record, because IBM cores are faster"

I never do what you do. I never say: "IBM's current TPC-C is not valid, because Oracle had faster response times" - would you accept this excuse from me? No? Why should I accept your excuse about SAP then?

IBM TPC-C slower response times:

http://www.c0t0d0s0.org/archives/6846-Arms-race.html

Me, I dont lie and FUD like you do, I accept that IBM has higher TPC-C. I dont deny that, like you do.

.

"....Now that is just pure FUD, you have no what so ever facts to match that statement. And IMHO it is clearly untrue. And shows that you don't know what you are talking about...."

So you dont agree that the IBM Mainframe cpus are dog slow? You have really drunk the IBM kool aid. I have many times, showed links on how slow they are. You want to see those links? Remember, I can ALWAYS back up my claims. I never lie nor FUD. I never make up new facts nor twist the truth - like you do. If you need six IBM POWER servers to match one Sun T5440 - how can that be FUD? Just read the white papers you linked to! I never make up things. I can always post links. You want to see links on IBM Mainframe cpus?

Clearly, you have not read those Mainframe links, so you dont know how slow they are. How can you say that I FUD then? You know NOTHING about the Mainframe CPUS and have not read any links on how slow they are - and still you say I FUD and that I state "untrue" facts. Great. This only proves that you FUD about me. You are a FUDer. Maybe there exist such links that I speak of? How can you then say I state untrue things - in effect, call me a liar? Someone is totally lost here.

.

"....BZZZZZ.. Wrong again.. Oracle had the 'clustered' TPC-C record, not the single Machine. Now they don't have the clustered record, it is also a POWER score...." When Oracle had the TPC-C world record - you know the vendor that pumped out the most TPC-C in the world, you know, fastest in the world - IBMers still claimed that IBM had the TPC-C world record, because "IBM's cores are faster". Great. Now, that is clearly an obvious lie. Who was fastest in the world? Oracle or IBM? Who could get the most TPC-C in the world? A hint: it was not IBM. How the heck can you say that IBM had the world record then? That is a lie.

.

"....You do know that POWER7 have been shipping for a long time don't you?..." It is not about POWER7. It is about how you lie, Jesper lille. I am proving that you lie. YOU claimed that the POWER6 is faster than Nehalem, beacuse it has faster cores. That is a lie, Jesper lille. Dont you understand? The question is: do Jepser lie or not? Answer: yes he lies. He claimed that POWER6 is faster, because the core is faster. I have mentioned this lie several times, and still you persist.

Ergo, Jesper lies. Never trust a liar.

.

"...Or is the only thing you can do whine FUD, come on grow up learn to research your arguments and do some logical reasoning. It's pathetic..."

Sure, prove that I FUD. Go ahead. Show me links, quote me. You have NEVER ever quoted me lying nor FUDing. In short, it is you, that FUD about me. This only proves again that you are a FUDer and liar

0
4
Thumb Down

Yeah right.. again again

".....evenue went down 13%"

Hey, I simply said you were right, Oracle does not have high margins on Hardware they are loosing money IMHO. Get over it. Don't go there if you are not prepared to face the findings.

"....Me, I dont lie and FUD like you do, I accept that IBM has higher TPC-C. I dont deny that, like you do.".

What sensible people like me are trying to hammer into the heads of people like you is that a cluster submission on the TPC-C benchmark is not a system.. it's a CLUSTER. Nobody ever denied that Oracle had the best clustered TPC-C benchmark submission. And that it was also the submission that had the biggest number of transactions...

But as it is a clustered submission.. then every sane semi skilled IT person should know that it's easily beatable by simply clustering together a few largish machines. Like IBM did with a 3 POWER 780'ies.

"...If you need six IBM POWER servers to match one Sun T5440 - how can that be FUD? Just read the white papers you linked to! I never make up things....."

Man ... you are dense and your argumentation is so flawed that it stinks.

Lets try to use your argumentation..

Yeah, the M9000 is twice as fast as the POWER 795.. and that is a fact just read the benchmarks:

POWER 795 specInt_2006rate -> 1440

M9000 specInt_2006rate -> 2590

http://www.spec.org/cpu2006/results/res2009q4/cpu2006-20091012-08891.html

So you need 2 POWER 795 machines to match one M9000, and that is a fact.

Only problem with this argument in the above example is that the POWER 795 used 4 chips out of 64 on the submission and the M9000 uses 64 chips out of 64 chips.

The real POWER 795 submission with 64 chips gives specInt_2006rate -> 11200

http://www.spec.org/cpu2006/results/res2010q3/cpu2006-20100817-12974.html

Kebbabert people are not laughing with you when you do math like that.. they are laughing at you.

And the whole mainframe CPU versus x86. When discussing this with you it's like discussing sex techniques with a virgin. You have absolutely no clue.

// Jesper

3
0

@Jesper Frimann

"...Nobody ever denied that Oracle had the best clustered TPC-C benchmark submission...." You are lying again.

Just as you denied that Intel Nehalem was faster than POWER6. That is also a clear and obvious lie, again. You claimed that even though you need four POWER6 to match two Nehalem, the POWER6 is faster - because the core is faster! That is clearly a lie and FUD. Intel Nehalem is faster, everyone except you IBMers understands it. How the heck can a POWER6 be faster, when you need twice as many POWER6 to match Intel Nehalem?? Ask Jesper Frimann: Which cpu is fastest for TPC-C? POWER6 or Intel Nehalem? Guess what Jesper would advise? I suggest you skip Jespers advices, they are wrong.

In the same vein, Oracle had the TPC-C record as clustered, yes. So what? Oracle was fastest in the world. If you needed the highest performance in the world, you had no choice but to use Oracle, And still you IBMers said "IBM still had the TPC-C record, because the POWER cores are faster". That is also clearly a lie. Who was fastest in the world? Oracle or IBM? Who is lying and FUDing?

.

"....Man, you are dense and your argumentation is so flawed that it stinks....So you need 2 POWER 795 machines to match one M9000, and that is a fact....Only problem with this argument in the above example is that the POWER 795 used 4 chips out of 64 on the submission and the M9000 uses 64 chips out of 64 chips...."

I didnt really get that. Do you mean that I compared six Sun T5440 servers with 64 cpus against one POWER server with 1 cpu? You mean I did an unfair comparison?

Let me recap: You need six IBM power servers with in total 28 of the super fast POWER6 cpus to match one Sun T5440 with four Niagara cpus in Siebel v8 benchmarks. How is that unfair to IBM? How is that FUD and "false statements"? Can you explain again? Or, are you just FUDing about me?

In your strange comparison of M9000 vs IBM 795, you compared 64 cpus vs 4 cpus - of course 64 cpus will win. That is unfair comparison. No one should compare more cpus against less cpus and claim victory - that is only something IBM would do. (IBM claims it is fair to compare 16 POWER6 cpus vs 4 Niagara cpus - because both have the same number of cores!!! If IBM tries to prove that POWER6 is faster than Niagara, then IBM must compare one cpu vs one cpu. Maybe one fast cpu has 100 slow cores? Then it is silly to compare core vs core. You need to compare the whole cpu vs cpu).

But when four Niagara cpus win over 28 POWER6 cpus - how is that unfair to the POWER6 cpus? I really dont understand how you think, Jesper. Seriously. That is maybe why you claimed that POWER6 is super fast - when in fact, it sucked badly. Come on, 28 POWER6@4.7GHz to match four Niagara@1.4GHz? Geez. How is that unfair to IBM?

.

"...Kebbabert people are not laughing with you when you do math like that.. they are laughing at you...." Actually, I never did math like that. You did. Guess who they are laughing at? You forget that I have a math degree, I would never think like you do.

.

.

"...And the whole mainframe CPU versus x86. When discussing this with you it's like discussing sex techniques with a virgin. You have absolutely no clue..."

Geez, Jesper. As I said, you dont have a clue about anything. As I have written earlier about the dog slow Mainframe CPUs (which IBM says are "the worlds fastest"):

Regarding the performance of Mainframes. A z10 with 64 cpus, give 28.000 MIPS. An 8-socket Nehalem-EX gives 3.200 MIPS - under software emulation. Software emulation is 5-10 times faster than native code. If Nehalem-EX could run Mainframe software natively, 8 x86 cpus would give 16.000 - 32.0000 MIPS. You need 8 Intel x86 cpus to match 64 Mainframe cpus.

http://en.wikipedia.org/wiki/Hercules_emulator#Performance

Also, another source, an Linux expert who ported Linux to mainframes, claimed year 2003, that 1 MIPS == 4 x86 MHz. Hence, a Mainframe with 28.000 MIPS corresponds to 112GHz. Pick a 8 core Nehalem-EX, which runs at 2.3GHz = 8 x 2.3 GHz = 18.4GHz. But, Nehalem-EX is much faster clock for clock, than Pentium 4. Whereas 1 MIPS then, is 1 MIPS today. Nehalem-EX may be 2 times faster than one Pentium 4, at same clock speed. Then those 18.4GHz of Pentium 4 MHz, corresponds to todays 36.8GHz. Again, you need just a few of Nehalem-EX to match 28.000 MIPS.

http://www.mail-archive.com/linux-390@vm.marist.edu/msg18587.html

Here we have another 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 Xeon. But todays modern server x86 cpus have 8 cores, which means they have in total 8 cores x 2 GHz = 16 GHz. Hence, we see that x86 at 16GHz is much faster than z196 at 2GHz. Again, you need only a few Nehalem-EX to match the biggest IBM Mainframe with 64 cpus.

.

In short, IBM = Master of FUD. Jesper, you have been fooled by IBM. You DO believe the IBM Mainframes have fast cpus. You DO believe the POWER6 is fast.

If you see any errors in my numbers, please point them out. But no weird comparisons please. I would appreciate if you just think a bit before you post. Remember, I have a math degree, whereas you have not.

0
2
WTF?

Damn you are scary...

"Just as you denied that Intel Nehalem was faster ...."

Yak Yak Yak... You are like a record stuck in the same loop. You never understood a word I said. That is clear.. and you just keep on repeating the same and the same again and again....I've never denied that Nehalem-EP was a faster CHIP than POWER6, but I've denied that it was a faster core.

Furthermore the scaling of Nehalem-EP was.. well.. EP like :)=

"That is maybe why you claimed that POWER6 is super fast - when in fact, it sucked badly."

Again in the myriads of benchmarks you have managed to find one multi-tiered Oracle controlled benchmark where you can claim a victory.

Now on Industry standard benchmarks like SAP2Tier, specINT_2006rate, specFP_2006rate, specJBB2005... POWER 570 wins with a good margin.. And even on your clustered, then each T5440 is 2.5 times slower than a stand alone POWER 570.

Wake up you level of fanaticism is scary.... Oracle is Great.. !!! All hail SUN I mean .. Oracle !!

"If you needed the highest performance in the world, you had no choice but to use Oracle,"

Again here you don't get it... real IT people who do sizing work and architect solutions know that if IBM submitted a 80 node POWER 750 node DB2 Purescale clustered submission, that this wouldn't make the POWER 750 the fastest machine in the World. It's just a node in a cluster.

" You forget that I have a math degree, I would never think like you do."

Jup, and you wrote your thesis in Bistromathic, which seems to be the only math you can use to prove your points.

"In short, IBM = Master of FUD. Jesper, you have been fooled by IBM. You DO believe the IBM Mainframes have fast cpus. You DO believe the POWER6 is fast."

Keb. I bear no illusions what so ever about how fast mainframe CPU's are. Both on a per core level and also on a per CHIP level. As a part of my Architect job at a Major CSI player I have access to NDA IBM sizing information. I've sized mainframe systems using that data and seen it hold water.

So, the CHIP isn't great IMHO when it comes to throughput, but they are nowhere as bad as you claim. And I couldn't give a damn how many amateur wannabe hackers you link to, it doesn't make it more real. Cause it's on the internet doesn't make it real.. hope you don't believe this guy is a great hacker just cause he's on the Internet:

http://www.youtube.com/user/NextGenHacker101#p/a/u/2/SXmv8quf_xM

Cause watching him hack, reminds me of someone...

// Jesper

2
0

@liar Jesper Frimann

"....I've never denied that Nehalem-EP was a faster CHIP than POWER6, but I've denied that it was a faster core...." Now you are lying again. Why do you lie so much? I wrote that POWER6 is slow, you need four POWER6 to match two Intel Nehalem on TPC-C and you claimed that POWER6 is actually faster, because it has a faster core. The Nehalem has 6 cores, and POWER6 has two cores. But you claimed that the POWER6 is faster.

Now you are lying AGAIN. Geez.

Who the heck cares about scaling? I have proven that you lie and FUD. I have showed that. On the other hand, you have NEVER proven me to lie or FUD. Everything I said, I can always back up to white papers, research papers, etc. Go ahead and prove that I am lying now. You have accused me of being a liar and FUDer, so I want you to prove it. Go ahead. Otherwise, stop FUD about me. Because that only makes you the liar and FUDer.

.

"...Furthermore the scaling of Nehalem-EP was.. well.. EP like..." Who the heck cares about scaling? I have not mentioned anything about scaling of Nehalem-EP. Why are you implying I have said something dumb about scaling, when I have not? It is like I say: "and regarding your earlier nazistic remarks, I dont agree with you". That would be a clear lie from me. But I dont lie, like you do. Why are implying I say weird things, and do weird math? I have not done anything of it. Is it just your ordinary normal FUD?

.

"...Again in the myriads of benchmarks you have managed to find one multi-tiered Oracle controlled benchmark where you can claim a victory..." Now you are FUDing AGAIN. You are beginning to getting tiresome, Jesper lille. How the heck did Oracle control the benchmarks from IBM and former Sun? Oracle did not even own Sun back then! Geez, Jesper. What is this shit you are doing? Can you stop lie and FUD, at least in ONE POST?

.

"...Again here you don't get it... real IT people who do sizing work and architect solutions know that if IBM submitted a 80 node POWER 750 node DB2 Purescale clustered submission, that this wouldn't make the POWER 750 the fastest machine in the World. It's just a node in a cluster...." No, it is you that dont get it. Let me ask you, if you needed the highest performance in the world, which company would you have to go to, IBM or Oracle? Oracle! How the heck can IBM and IBMers claim they are still fastest in the world, because "IBM cores gave 4.7 times more tpmC per core"?? That is an outright lie, that technically ignorant executives might believe. FUD and lies, again. This is a lie, Jesper. Dont you see the lie?

IBM would never get a high record with such a configuration of 80 nodes, if IBM had, they would have immediately published it.

Say after me: "Oracle had the top spot - that means they are fastest in the world."

.

"....Jup, and you wrote your thesis in Bistromathic, which seems to be the only math you can use to prove your points...." Actually, you dont prove your points at all. When you try, you get muddy and not logical.

.

"...And I couldn't give a damn how many amateur wannabe hackers you link to, it doesn't make it more real...."

Regarding the "super fast Mainframe cpus". They are dog slow. Them cpus are a derivative of POWER6 cpus - that is the reason they are dog slow. They have few cores, and run at 5GHz or so - typical POWER6. If IBM based the Mainframe cpus on POWER7, they would be good. Because POWER7 is a good cpu, probably the fastest in the world right now for some work loads. For other work loads, Oracle T3 is the fastest in the world right now.

IBM Mainframe z196 is far far far behind. One CPU has almost 300GB cache (L1 + L2 + L3 + L4) and use 5GHz speed, and STILL it is only 50% faster than the previous generation z10 Mainframe cpu. Can you imagine that? 300 GB cache???? One CPU??? That is hilarious. And proves the z196 is an abomination that should have been killed off and never left the laboratory.

As I explained earlier, high clock speeds will be severly punished if you do a cache miss. That is the reason one cpu has closer to half a TeraByte of cache. What a failure.

Regarding those "amatuer wannabe hackers" you talk of:

-One of them wrote an IBM Mainframe emulator running on x86 - hardly a amateur, eh?

-The other ported Linux OS to IBM Mainframe and could compare software running on x86 and on Mainframes - and make a comparison. Does he sound like a amateur?

-And the last link I showed, was to Microsoft, they hired "Forrester Research Inc" to do a comparison between x86 and Mainframes, and also hired and independent third-party performance benchmark, audited by Meta Group. Are they also amateurs?

In short, when I show links or benchmarks, you immediately dismiss them as FUD, lies, and amateurs. When you show links or benchmarks, I accept them. What does that tell you, about Jesper Frimann?

.

You make it difficult for me. Sometimes you say one thing, the other time you say the opposite:

SIEBEL:

Oracle: high throughput, bad response times

IBM: low throughput, good response times.

Jesper's conclusion: IBM wins because of good response time.

TPC-C:

Oracle: low throughput, good response times

IBM: high troughput, bad response times

Jesper's conclusion: IBM wins because of good throughput

Fantastic. How can Oracle win against such a fanatic bias? And you call ME a fanatic? At least I am objective, and let the white papers, benchmarks decide who is fastest. You let your murky fanatic judgement decide who is fastest, and in every case it is IBM. Even when IBM looses, IBM is fastest. And I am the fanatic? More FUD and lies from you?

0
1
Thumb Down

Yeah Yeah Yeah..

"Who the heck cares about scaling?"

Well I do.. Scaling is a very important. Which is why it is something that should be looked at when comparing systems and processors. Specially if you are comparing Itanium/POWER/SPARC64 versus x86. Cause Itanium/POWER/SPARC64 will do from 4-256 cores with the same chip. Intel x86-EP will only natively go to 2 chips and 8-12 Cores. Hence when you pull out your Nehalem-EP numbers.. then they are only valid for workloads that does not require more scalability than the Nehalem-EP chip can deliver. For larger workloads you have to look at Nehalem-EX. It's actually pretty simple.

If I need 300 specINT_rate2006, then it's kind of stupid to be looking at a Nehalem-EP x5570 processor cause it'll only take you to about 266. Where as the nehalem-EX X7560 will take you to.. well.. 1400 or so..

So that whole scalability comes at a price, and that is the per core throughput. Again the two above processors in 2 socket configs:

http://www.spec.org/cpu2006/results/res2010q3/cpu2006-20100621-11923.html

and

http://www.spec.org/cpu2006/results/res2010q1/cpu2006-20100315-09857.html

That is 266 and 33.25 per core specINT_rate2006 for the Nehalem-EP and 385 and 24,06 per core for the Nehalem-EX. So the 45% increase in chip throughput comes at a price of a 38% drop in per core throughput.

And you talked about TPC-C

Then

Nehalem-EP (Submit date 04/08/10 ) 631766 tpmc and 78971 tpmc/core:

http://www.tpc.org/results/individual_results/HP/HP_DL370_G6_OEL_TPCC_ES.pdf

And

Nehalem-EX (Submit date Aug 27, 2010) 1807347 tpmc and 56489 tpmc/core

http://www.tpc.org/results/individual_results/HP/HP_ProLiant_DL580G7_2.26GHz_es_100830_Energy_v2.pdf

And finally

POWER6 (Submit date May 21, 2007) 1616162 tpmc and 101010 tpmc/core

http://www.tpc.org/results/individual_results/IBM/IBM_570_20070522_ES.pdf

Again Nehalem-ex is 40% slower per core than -EP. And the picture is clear POWER6 is faster per core, (on this benchmark) than Nehalem.

And also notice that the POWER6 submission is from 2007, not the fastest POWER6 made, and the Nehalem are using SDD drives (which IMHO is a pretty big factor).

Now a current POWER7 system like this one:

http://www.tpc.org/results/individual_results/IBM/IBM_780_TPCC_20100719_es.pdf

does 1200011 tpmc and 150001 per core. Which is almost x3 per core of Nehalem-EX and almost twice that of Nehalem-EP per core performance.

And the sheer fact that you can come with such a statement, only shows that you still have much to learn.

"No, it is you that dont get it. Let me ask you, if you needed the highest performance in the world, which company would you have to go to, IBM or Oracle? Oracle! How the heck can IBM and IBMers claim they are still fastest in the world, because "IBM cores gave 4.7 times more tpmC per core"?? That is an outright lie, that technically ignorant executives might believe. FUD and lies, again. This is a lie, Jesper. Dont you see the lie?"

I wouldn't go to Oracle that's for sure. It's not the fastest machine.. it's a cluster .. CLUSTER. And if you read the pricing information on their TPC-C benchmark you'll see that you don't even buy the software... you lease it.. For the exact amount of years that the benchmark has to do TCO on. You have to pay the listed amount of money every 3 years... And there is NO upgrade protection.

So who is trying to Bull who ? On the POWER 595 benchmark you at least buy the software and then only has to pay (which can be expensive enough btw.) software maintenance and support.

So buying 2 years of extra support for DB2 on the (going from 3 year -5 year TCO) would cost you ... 203,827 USD-Discount. On the Oracle solution using the pricing scheme that they use on the benchmark it would be.... 7,872,000 USD for 3 years.

Now who is b*llsh*tting who ?

You are so drunk on the Oracle/SUN Koolaid that you don't care to read the fine print.

"In short, when I show links or benchmarks, you immediately dismiss them as FUD, lies, and amateurs. When you show links or benchmarks, I accept them. What does that tell you, about Jesper Frimann?"

That I'm much much better at coming up with links than you are ?

It's not my fault that IBM doesn't release any benchmarks on Mainframes. Go complain to them. I just know the numbers that we use. Would a 8 core Nehalem-EX chip faster than a 4 core Mainframe chip on benchmark like specINT_rate2006. Jup sure no argument from me.

But 8 times faster, is so far off the target that it can only be described as FUD, or well.. stupidity.

Also here the type of workload plays a big part. Mainframe Cores have never been known for their ability to crunch numbers, Moving data on the other hand they are pretty good at, and that is also one of the secrets why they can run at such high utilization. Which is also a factor that you have to look at in RL.

"That is hilarious. And proves the z196 is an abomination that should have been killed off and never left the laboratory."

You really really don't get it. Your fanaticism is scary.

"You make it difficult for me. Sometimes you say one thing, the other time you say the opposite.."

That is most likely cause you don't understand what I say.

Now as for response time. I've only pointed out differences in that on the benchmark and said that you can trade response time for throughput.

On the TPC-C benchmark you are quite right there is a big difference in response times, between the POWER 595 and the T5440 clustered benchmark.

On that particular benchmark then one of the key factors is surely the use of SDD on the T5440, which wasn't available back when the POWER 595 benchmark was made. It's a pretty big difference.

But the T5440 clustered benchmark do seem to have idle processing power that could have been used to increase the throughput at the cost of response time. How much is hard to say, why well according to a friend of mine who's a certified Oracle RAC dude, then it's cause they are pushing the limits for scalability on the benchmark, hence they aren't hitting the optimal per core throughput.

// Jesper

0
0

Nice

A balanced article and an interesting viewpoint. One can hardly expect IBM to announce bleed figures to the industry, so you're right - the question is why aren't HP and Oracle publishing their own takeout figures to balance the account.

Maybe:

(a) Hurdgate is taking up all their PR resources

(b) There aren't any takeouts of which to boast (or embarrassingly few)

(c) There's an announcement coming that will render takeouts a moot point

I leave it up to loyal Reg commentards to fill in options (d) through (z)...

0
0
Bronze badge

And that...

You can always get a better deal migrating to another supplier.

3
0
Anonymous Coward

Customers Primary Concern

Not really, what matter in enterprise sales is the long-term support. You don't want to risk your business for a little saving from migrating to another supplier. That's why these machines are usually called as mission critical or business critical machines. And that's why Hurd's scandal and Sun's acquisition matters for customers.

On the other hand, sometimes, keep quite is better for the companies. That's probably why HP and Oracle don't say a single thing. Everyone can say anything he wants, but in the end the real fact data (read - financial report) can't be denied.

0
0
Unhappy

Lucky you...

Either you have no management, or they have a brain. Mine is wooed by glossies and free lunches. They might save a buck, but it's usually for a product that doesn't fit the bill, and usually requires more labor hours to implement and/or maintain. It's a vicious circle of flip flopping with you holding the bag as their latest preferred hardware line goes down...

0
0
FAIL

Those servers

...appear to be tape libraries.

1
0

Marketing talks...

Competitors in fact are right on refusing to comment about IBM's takeouts propaganda. If they answered, it would just make its argument stronger.

The clients know well this is sales talk and thus not worth to listen too closely.

I would like to see Big Blue telling how much x86 systems they were able to convince to move into Power/AIX lock-in formula. It should be as unlikely as seeing Linux running on Power servers with no AIX system at all....

2
0
Stop

Out of frying pan,. into the fire

Unfortunately, the vast majority of IT execs are not looking beyond Unix to see how they can reduce the costs of doing IT business.

There continue to be reasonable and very valid needs for a Unix architecture, but for the majority of applications, the migration to an x64 industry standard platform is now more valid than ever.

With this comes not only lower server hardware, but lower costs of software, management, peripherals and staff availability/rates.

0
0
Joke

Re: Out of frying pan,. into the fire

"Unfortunately, the vast majority of IT execs are not looking beyond Unix to see how they can reduce the costs of doing IT business.

There continue to be reasonable and very valid needs for a Unix architecture, but for the majority of applications, the migration to an x64 industry standard platform is now more valid than ever.

With this comes not only lower server hardware, but lower costs of software, management, peripherals and staff availability/rates."

Are you a politician or a sales man? ;)

0
0
FAIL

Re: Pure FUD

Typical Kebabbert hyperbole...

"Oracle have high margins, and at the same time talk about IBM? That must be the weirdest thing I have heard. IBMs Mainframes takes 10% of all money of the entire server market, and number of Mainframe shipments is down to 0.01% of the entire server market. Now THAT is what I call a fat margin. 0.00% of all servers, take 10% of all money. This, I read here on theregister.co.uk."

How about links to actual articles? There are thousands of articles on the register each year.

The numbers are too precise. Sounds like you made them up.

"Also talking about "questionable value proposition" - you needed 6 (six) IBM Power P570 servers to match ONE Sun T5440 server in SIEBEL v8 benchmarks. And ONE P570 server costed $413.000 USD, whereas one T5440 costed $76.000 USD. You needed IBM servers worth of 6 x 413.000 = 2.5 million USD to match one Sun server. Now that is what I call questionable value proposition from IBM."

"...Again you keep repeating that story again and again.. it's 2/3 filled up POWER 570 and a 1/8 th filled up POWER6 p570 versus one T5440. Hence your claims are ... false.. and stink of FUD..."

"Ok, I did not know that there is a difference between IBM p570 and IBM POWER 570 - I thought it was the same server. So, I should instead have said: "you need six p570 and POWER 570 servers to match ONE Sun T5440 server on official SIEBEL benchmarks v8.0" - then my statement would have been correct. So, I dont see how I FUD? Do you deny that you need six IBM servers to match one Niagara server?"

Jesper is so right. You repeat the same crap over and over again. You keep ignoring the fact that both p570 and power 570 have been replaced by the power 7 based 770.

You need to come up with new benchmark comparisons because these are old and only interesting to historians.

"Not to mention those uber slow Mainframe CPUs. One of the newest z196 Mainframe CPUs, which IBM calls "worlds fastest CPU" at 5.2GHz and 300MB cache (L1 + L2 + L3 + L4) is really dog slow. You need 5-10 of the z196 cpus to match one modern x86 cpu. How can IBM say that the z196 is the "worlds fastest cpu"?"

Oh, really. Where is your data for that 5-10 claim. Oh, right you are too lazy to go find perf data on your own, and you just keep repeating the same numbers over and over again. At least you changed z10 to z196.

"Oh wait, of course, now I understand. IBM and IBM supporters have a weird logic, that is the reason!"

"Fastest" is always subjective. It's pretty clear the IBM focused on clock speed. Get over it. Everyone else has moved on already.

"Earlier, when Oracle had the TPC-C world record, IBM said that because IBM has faster cores, the TPC-C world record still belonged to IBM. If you looked at no 1 on the TPC-C list, you saw Oracle. But still IBM said that they had the record, because they "used faster cores"."

TPC-C rightly classifies submissions as clustered, or not clustered. Oracle used multiple machines to get to number 1. You made it about cores. Everyone tells you that core counts are important for software licensing. Then you make the same old exaggerated statements without any facts to back them up. Then a week a later you quote your own posts and say that your exaggerations must be true.

"Or, the POWER6. You need four POWER6 cpus to match two Intel Nehalem - and guess which cpu is the fastest? POWER6 OF COURSE! Because "the POWER6 core is faster" - yes that is true! They say that! Weirdos."

Guess what!? Update your numbers. It's Westmere vs Power 7 now.

0
0

Coward,

Hey coward, you are also spreading the IBM FUD I see.

"How about links to actual articles? There are thousands of articles on the register each year."

You answered your question yourself. There are thousands of articles, I dont have time to read through those especially not for a coward IBM FUDer like you. But, it was an article about IBM Mainframes, and most probably I have written a post there, on how dog slow the Mainframes are. You know, the Mainframe cpus are a derivative of POWER6. That explains why they are so slow cpus. So, if you really want to find those numbers (which I remember, because I got shocked when I saw them) I quote, you now have plenty of information to find that article.

.

"....Jesper is so right. You repeat the same crap over and over again. You keep ignoring the fact that both p570 and power 570 have been replaced by the power 7 based 770. You need to come up with new benchmark comparisons because these are old and only interesting to historians...." No, they are not interesting only to historicians. Those benchmarks prove that Jesper Frimann lied and FUD. (And continues to do that). Of course I know that those machines are old, but that is irrelevant. The relavant thing is that I prove that Jesper lies. I have now established that Jesper lied, and that I was correct all the time.

When Jesper says I "FUD" and use "false statements" about the SIEBEL benchmarks, because I compare one Sun server to six IBM servers - everyone can see that it is not true, I do not FUD. Everything I said, is written in the white papers, just read them! You NEED six IBM servers to match one Sun server. It is in fact, Jesper that FUDs and lies about me, and about Sun servers. How can this statement be FUD and false? "You need six IBM power servers to match one Sun server in Siebel v8 benchmarks according to those white papers". That statement is correct, and has always been.

.

".....Oh, really. Where is your data for that 5-10x claim. Oh, right you are too lazy to go find perf data on your own, and you just keep repeating the same numbers over and over again. At least you changed z10 to z196...."

See the post above, there I show that one modern x86 cpu is 5-10x faster than one IBM Mainframe CPU. And of course I changed z10 to z196. The z196 is only 50% faster than the old z10. So the numbers are 50% better in IBM's favour. But that is not much.

.

"...."Fastest" is always subjective. It's pretty clear the IBM focused on clock speed. Get over it. Everyone else has moved on already...." Bull sh*t. IBM lies when they say in press releases that z196 is "the world's fastest cpu". I show above, that it is dog slow cpu. Right below Jesper's unstructured ramblings.

.

"...TPC-C rightly classifies submissions as clustered, or not clustered. Oracle used multiple machines to get to number 1...." Who cares about who is fastest? That is not the point. The point is that I prove that you should not trust IBMers - because you lie. How can IBM claim they still have the world record in TPC-C because "one core was faster"??? Now that is a lie.

.

"...You made it about cores. Everyone tells you that core counts are important for software licensing..." Who cares about software licensing prices? We talk about which cpu is fastest. POWER6 or Niagara. And guess what? IBM compares 32 POWER6 cores to 32 Niagara cores, and concludes that the POWER6 is faster cpu. But, what IBM does not reveal is that IBM compared 16 POWER6 cpus vs 4 Niagara cpus - of course 16 cpus are faster than 4 cpus!! The point is - IBM FUDs and makes unfair comparisons! IBM claims that POWER6 is faster, but it is a lie.

I reminds me of when Microsoft in a study compared Linux vs Windows TCO, and concluded that Windows is cheaper. What MS did not reveal, was that Linux ran on an IBM Mainframe, and Windows on a PC - of course Windows is cheaper then! But still, that is unfair and a false statement to claim that Windows has lower TCO than Linux.

.

"...Then you make the same old exaggerated statements without any facts to back them up. Then a week a later you quote your own posts and say that your exaggerations must be true...."

Another lie from you. Quote me, when I do like that. Prove your false claim. You can not prove it? Oh, please stop FUD about me, then. Right now, it is you that FUD, about me.

.

"....Guess what!? Update your numbers. It's Westmere vs Power 7 now...." Who cares about Westmere vs POWER7? My point is that Jesper LIED. Jesper lied when he says that POWER6 is faster than Intel Nehalem - because one core is faster.

I am proving that Jesper lies, you should not trust Jesper's posts. That is why I wrote that. Did you not understand it?

On the other hand, both you and Jesper claims that I lie and FUD. Go ahead, quote me when I lied or FUDed. Prove that I posted lies. You can not? Stop call me a FUDer, then.

Remember, I can always back up my claims with white papers, benchmarks, research papers, articles, etc.

0
3
Happy

I think you need to check...

Your blood pressure Keb.. All that Marketing B*LL is full of Salt, bad cholesterol and female hormonal like substances. It's not good for U.

You can call me a liar or a fudder as much as you want, it doesn't make it more right.

// Jesper

1
0

@liar Jesper Frimann

"...You can call me a liar or a fudder as much as you want, it doesn't make it more right...."

Look, I have just PROVED that you FUD and lie. I have established that you do. The evidence is there. For instance, you say "even though you need four POWER6 to match two Intel Nehalem, the POWER6 is fastest". This clearly a lie.

You claim that I lie and FUD, go ahead and prove that. If you can not, it is you that FUD about me. And as I have already proved that you FUD, most probably this is just more FUD from you.

0
0
(Written by Reg staff)

Re: @liar Jesper Frimann

Can you take this outside now, please? Thanks.

0
0
Troll

*BING* You just used the word FUD for the 1000th time.

And your price will be ... nothing.

Perhaps a reading lesson would be in place ?

Some quick googling.

http://forums.theregister.co.uk/post/916032 this post in this thread:

http://forums.theregister.co.uk/forum/1/2010/11/10/amd_opteron_server_roadmap/

"Now I wouldn't argue against that the Nehalem-EP chip at 2.93GHz with 4 cores is faster than a POWER6 Chip with 2 cores. The POWER6 core is still faster than Nehalem. And POWER7..."

Again.. your only defence left is crying FUD FUD FUD.. when your arguments don't hold up.

// jesper

0
0

@Jesper Frimann

You have several times accused me of using false statements and FUD. Prove that I lie and FUD, I have asked you many times, and you have never proved it.

I quote BMSEER, yes. But BMSEER never makes the numbers up, he quotes from IBM web sites and Sun Web sites, and quotes white papers and official bench marks. How can official white papers be FUD, just because BMSEER links to, and quotes them?

.

.

Regarding your statement about "POWER6 is faster than Intel Nehalem, because one core is faster" - I have googled a bit but can not find that statement. I dont understand. Where is it?

In this post

http://forums.theregister.co.uk/forum/1/2010/01/20/ibm_power7_q1_launch/

at time stamp "Posted Thursday 21st January 2010 13:41 GMT"

I write many things and quote web pages, about how slow the POWER6 is. The structure of my post is something like this:

1) ...

2) ...

3) ...

4) "Also IBM really needs POWER7 to be very good. POWER6 is not very impressive despite its crazy specifications. For example, a 4 CPU POWER6 4.2GHz machine (total 8 cores) performs similar to a 2 CPU Nehalem 2.93GHz machine (also total 8 cores) in TPC-C. Although POWER6 is older but Intel is closing very fast and that really worries IBM."

5) ...

6) ...

.

Jesper replies to 1) 2) 3) and 5) and 6). But there is no answer to 4). I dont get it? What happened to your answer, Jesper? Where is your text about 4)? Why does your text only answer 1) 2) 3) 5) and 6) but not 4)??? Where is that part that answers 4)???

.

.

Further down "Posted Monday 25th January 2010 10:14 GMT", I talk about your answer to 4). But first I complain on something messing up my text in my post. I write, regarding your answer to 4) that your answer is flawed:

"It is like TPC-C bench. We see that four POWER6 is needed to match two Nehalem CPUs, and what is your conclusion? THAT POWER6 IS FASTER, BECUASE IT HAS FASTER CORES!!! You start to ramble about how many cores, etc. But it is not relevant how many cores when we talk about performance of CPU vs CPU. If we talked about performance of cores, then yes your objection is relevant. But now we are talking about CPU vs CPU! So answer me this; how many POWER6 is needed to match two Intel Nehalem on TPC-C? The answer is four. So how can POWER6 be faster than a Nehalem? Your logic is a bit strange?"

Here I ask you questions on your answer to 4), but you dismiss my question, realizing you did a blunder when you claimed that POWER6 is faster, because of faster cores.

.

And still further down "Posted Monday 25th January 2010 12:22 GMT", I again ask you the same thing because you never replied:

"I know you are reading this. Can you answer to my question? If you need four 5GHz POWER6 to match two 2.93GHz Intel Nehalem on TPC-C, how can you justify your conclusion that the POWER6 is the faster CPU? If I can only choose one CPU, which should I choose then? Which is fastest? I do not talk about cores, nor pricing of cores. I talk about performance of the whole CPU."

Still you didnt answer.

Where is that part where you wrote that "POWER6 is faster than Nehalem"? As soon as I see your dumb answer, I immediately grab the opportunity and are at you, again and again. That is a blunder I can not miss.

Can you edit posts somehow? What happened? In one of my posts in that thread, I complain about something messing up my post. Has it messed up your post too, or have you edited your text afterwards?

WTF?

0
0
Thumb Down

Ok it is now official...

You are mad.. simply mad.

Have it ever occurred to you that the reason you cannot find any link to me saying that the POWER6 CHIP will always do more throughput than a Nehalem-EP or than an T2+ Niagara is cause I have constantly written CPU core or core. Admit it you have misunderstood me and get over it. Look... I have constantly tried to list the facts.. Good example here:

http://forums.theregister.co.uk/post/672443

I list the performance difference between T2+ and POWER6 at equal Thread count, equal Core count and Equal Chip count. You don't get more honest than that ?

So perhaps you should duck out of the way of the Orbital Mind Control Laser next time it passes.

// Jesper

0
0

@liar Jesper Frimann

And now I want you to quote me, when I lie and FUD - as you many times have claimed. You have claimed that in numerous posts. Go ahead now and quote me. I have quoted you in two different cases where you lie and FUD. I can quote you many more times.

Now I want to see posts where you quote me. Go ahead. Prove that I lie and FUD as frequently as you claim. Go on. I am waiting.

I have asked you this, many times but never have you quoted me on lying and FUDing. If I really did FUD and lie as much as you claim, then it should be easy to show lots of posts. So, where are they? Quote me. Oh, you can not find those numerous quotes where I lie and FUD all the time?

0
0
Linux

Fud it is...

"Now I want to see posts where you quote me. Go ahead. Prove that I lie and FUD as frequently as you claim. Go on. I am waiting."

Woooo... that was an invitation I just couldn't resist.. problem is that it's hard to know if you are just ignorant sometimes or just a really really bad FUDSTER.

http://forums.theregister.co.uk/post/676357

"One Mainframe z10 CPU gives you 437 MIPS in native code. Software emulation is a factor 5-10x slower. One large Mainframe with 64 cpus give you 28.000MIPS:

http://en.wikipedia.org/wiki/Hercules_emulator#Performance

You need less than 16 Nehalem-EX CPUs to match 64 Mainframe CPUs. "

First of all A z10 Mainframe has 4 MCM modules each with 4 CPU's which each has 4 CPU cores. That gives a total of 64 CPU CORES. (there are more but those aren't used for processing so it's 64 for Actual workloads) NOT CPU's, hence you are either deliberately or by sheer ignorance doing the whole math wrong comparing CPU cores with CPU's. A Mainframe CPU does 1750 MIPS not 437, that is the core... CORE.

So either you are really spreading FUD or you simply just don't know what you are talking about.

Lets see what else we can find:

http://forums.theregister.co.uk/post/672926:

"Also, I have heard that POWER7 is basically a couple of stripped down POWER6. But you claim it is more similar to a...."

Now calling a 'FAT' Out of order 8 Core CPU a 'couple of stripped down POWER6'es is.. well just throwing mud at the 'competition'. Specially when POWER7 CHIP delivers someting form 4-6 times the throughput of a POWER6 CHIP.

http://forums.theregister.co.uk/post/570575:

"And you know that a high clocked CPU as the Power6 uses lots of power. Maybe 400watt? 500watt? "

Now that is a good piece of FUD. A POWER 560 with 8 Chips burns with MAX ram MAX disks MAX Adapters MAX CPU's 2,246 Watt at 100% utilization.

Now according to your calculations then just the CPU's should burn 3200-4000 Watt.

Try it our yourself here:http://www-912.ibm.com/see/EnergyEstimator. The right word for such statements are either ignorance or FUD.

Should I continue or have you had enough ?

// Jesper

1
0

@Jesper frimann

"...A Mainframe CPU does 1750 MIPS not 437, that is the core, CORE. So either you are really spreading FUD or you simply just don't know what you are talking about...."

Ok, so how is this FUD from me? Can you explain again? I didnt understand that.

According to wikipedia, a z10 Mainframe which has 64 cpus, gives 28.000 MIPS. This means a Mainframe cpu gives 28.000 / 64 = 437 MIPS. Ive read somewhere that IBM claims a fully loaded z10 Mainframe with 64 cpus, gives ~30.000 MIPS. Now, 28.000 MIPS and 30.000 MIPS are almost the same. IBM confirms what wikipedia claims.

But you claim each Mainframe cpu gives 1750 MIPS. This means that a Mainframe with 64 cpus would give, 64 x 1750 = 112.000 MIPS. This huge number contradicts wikipedia and IBM.

Dont you see that something is logically wrong? IBM and wikipedia both say the same thing, you say something else. Now how have you proved that I FUD? We can derive contradictions from your claims, because your claims are not logically sound. Something is clearly strange, which you dont notice. In this case, it is you that is again vastly exagerating IBM's performance.

If anyone is FUDing here about IBM's performance, it is you. I have earlier showed you this calculation and quoted wikipedia, and still you claim that I FUD and make up this number of 437 MIP and never post link? It proves you are FUDing about me.

.

"...Now calling a 'FAT' Out of order 8 Core CPU a 'couple of stripped down POWER6'es is.. well just throwing mud at the 'competition'.... Specially when POWER7 CHIP delivers someting form 4-6 times the throughput of a POWER6 CHIP."

First of all, I am clear it is hear say, I wrote "I have heard that..." It is not a claim I make. The nature is more of a question of the hear say to be confirmed from you. I am not making a claim here, which can be false/true.

"I have heard that Michael drunk yesterday, can you confirm this?" - a question. Not a true/false statement.

"Michael drunk yesterday!" - a true/false statement

It is a big difference between declarative statements or stating a question. If you dont know this, it is because you have not studied logic. I suggest you do. Your logic is somewhat flawed, because we can derive contradictions from your claims. That is a definition of "flawed logic", you show us.

Regarding that hear say of IBM power7 (I am clear about it is hear say, it is not a declarative statement - it is a vast difference!), if you take four dual core POWER6 and glue them together, then one POWER7 with 8-cores gets "4-6 times the throughput of a POWER6 CHIP". Is that not true? How is this hear say I have read, FUD? How can that be FUD? It makes logically sense! You take four POWER6 and glue them together into a POWER7 - then you get 4x better throughput. It makes sense. How can you claim this is not reasonable and call it FUD???

.

"...Now that is a good piece of FUD. A POWER 560 with 8 Chips burns with MAX ram MAX disks MAX Adapters MAX CPU's 2,246 Watt at 100% utilization...."

Again, I am ASKING something that waits to be answered. If you see the question mark, it is clear it is a question. Again, if I hear something, I clearly state that I have heard it. It is a big difference between hear say and a declarative claim. Claims can be true/false. Hear say can not be true or false - they need to be confirmed.

After you have explained that POWER6 does not use that much power, I have stopped asking that question. You have never seen me talking about POWER6 power usage again. It is because you answered that question. If you explain something to me, I stop ask that at once.

If I explain you are wrong on something, you clearly continue to FUD. You can claim the Mainframe cpus are fast, even though you confess they are slow, and even though you have seen my links that show how slow they are. This is clearly FUD: both you and me and Mainframe experts agrees the Mainframe cpus are dog slow - but still you insist they are fast and that I FUD and make false statements about the Mainframe CPUs. This is clearly FUD about me.

.

"....Should I continue or have you had enough ?..."

Please continue. You have not proved anything yet, you have not showed anywhere that I FUD or make false statements. Maybe it is because you dont understand the difference between declarative statements and questions and hear say - we are having this mis communication? I claim something which is logically correct, which you wrongly interpret as FUD.

On the other hand you clearly FUD and write false and contradictory statements - but maybe you did not mean to form them as declarative statements?

.

.

This is taken from the post below.

I write:

"...There are numerous other weird claims from him...."

You answer:

"...Perhaps that is cause you don't understand them. Yes I am known for thinking out of the box, and coming up with creative solutions to traditional problems...."

Jesper, I have proved that your logic is not sound and it is contradictory. That is not a sign of you "think out of the box". It is sign your logic is flawed. Maybe that is the problems we have in communicating? I use correct logic, which you interpret as FUD. And when you use flawed logic (which you dont understand is flawed) I react and claim you are claiming false statements - which you do from a logical view point.

How about that? (Please observe that this is a question from me that needs to be answered, not a declarative statement which can be false/true)

0
0
(Written by Reg staff)

Re: @Jesper frimann

OK - it's a new week, and I'm not going to referee this particular playground scrap any longer. Could you find a way to agree to disagree? If not then I'll start rejecting with fist of iron. Ta.

2
0

@Sarah Bee

I dont get it. Who are you that talk about "rejecting"? What are you going to reject? Posts? Are you a moderator? Why is your nick "Sarah Bee" then, and not "Moderator"?

Can you explain?

0
0
(Written by Reg staff)

Re: @Sarah Bee

I'm a moderator, yes. You'll see I've got a Reg badge.

I'm just asking you both to reel it in a bit because you've been going at it hammer and tongs for a while now and it's time to take some deep breaths and turn your attention to something else, for all our sakes. Ta.

0
0

@Sarah Bee

Oh, I didnt know that you are a moderator. Sure, if you say so.

(I just can not stand the FUD and Sun and SPARC bashing calling it slow when it holds several World Records - then I react)

0
0
(Written by Reg staff)

Re: @Sarah Bee

Yes, I know, arguments happen, but eventually you have to let 'em go.

Also the constant shouting of 'FUD' is hurting my ears.

0
0
rch

To the lunatic

Kebabbert

# for instance, you say "even though you need four POWER6 to match two Intel Nehalem, the

# POWER6 is fastest". This clearly a lie.

Strange. I have never seen Jesper make such a blanket statement. In fact when googling it I only find Kebabbert, like a lunatic stalker, calling Jesper a liar and a FUDer all over the internet.

Why don't you provide one of your many links and back that claim up?

1
0

@rch

I cant find it.

But I have seen Jesper meka such a blanket statement. That is not an extraordinary statement from Jesper.

He claims many weird things, things like... "When T2 got a speed bump from 1.4GHz to 1.6GHz, it can be considered as a next gen cpu - and should be pitted against next gen cpu from IBM: POWER7".

To that I objected and explained that T3 is the next gen cpu, that you can not count a speed bump as a next gen cpu, but he persisted.

.

There are numerous other weird claims from him.

I think it is quite fun that his statements are so extraordinary remarkable that people question the truthness in his statements! That is clearly a sign of how strange his statements are.

Maybe I should start to post links and quote him everytime he makes a strange statement - which is quite normal for Jesper. Because people doubt that he really said what I quote.

What strange statements from Jesper do you want me to quote?

When he claims that POWER6 is faster cpu than Niagara cpu because POWER6 has a faster core? Yes, it is true. He said that! No matter how many cores Niagara has - if the core is slow, then the entire cpu must be slow. According to Jesper. Even if Niagara has 10000 slow cores, it can not be faster than POWER, because one core is slow. Remarkable statement from Jesper.

0
2
Headmaster

Try to understand this then...

"When T2 got a speed bump from 1.4GHz to 1.6GHz, it can be considered as a next gen cpu - and should be pitted against next gen cpu from IBM: POWER7".

Please don't use pseudo quotation marks on a statement that I have clearly not said, if you are not familiar with the use of quotation marks don't use them.

With regards to T2 versus T2+, some call it a speed bump, others actually lists it as a separate processor. But if I am sooooo wrong why then does the SPARC WIKI list's the T2 and the T2+ as a separate processors. Try to have a look:

http://en.wikipedia.org/wiki/SPARC

Now is that Wiki written by SUN/Oracle hostile bad guys who want to bring down the world ?

Not likely.

POWER6+ was also just a speed bump, but it's still an, to use a Intel term, *tick* *tock* generation. Just as Westmere-EP is to Nehalem-EP.

"There are numerous other weird claims from him."

Perhaps that is cause you don't understand them. Yes I am known for thinking out of the box, and coming up with creative solutions to traditional problems.

"I think it is quite fun that his statements are so extraordinary remarkable that people question the truthness in his statements! That is clearly a sign of how strange his statements are."

Since when did you start to become plural ? My claims are normally always backed up by 'facts'.

Not that I'm not wrong sometimes cause I am.

"When he claims that POWER6 is faster cpu than Niagara cpu because POWER6 has a faster core? Yes, it is true. He said that! No matter how many cores Niagara has - if the core is slow, then the entire cpu must be slow. According to Jesper."

Try to read what I wrote:

http://forums.theregister.co.uk/post/672443

I list the performance difference between T2+ and POWER6 at equal Thread count, equal Core count and Equal Chip count. You don't get more honest than that ?

And my conclusion was:

"So basically Niagara has a slim lead on POWER6 only when comparing chips to chips, and at ithe best result is a 60% lead and that is only against lower clocked POWER6 on SPECint_2006rate. On specfp_2006rate (16 way) each 8 core, 32 threaded T2+ chip is only 20% faster than 2 core 4 threaded POWER6+ chip."

If you had bothered looking after a post where I get enough of your rant and lash out at your arguments then you should have quoted this one from (http://forums.theregister.co.uk/post/673128) :

"The whole point is that it doesn't suck, power6 is damn fast. The project I am responsible for has no less that 16 power 570'es and a few p595's. And damn they are fast, and actually quite forgiving due to their high Ghz. Again if there is a CPU core that sucks then it is the T2. Now both Itanium and POWER has managed to keep One of the reasons why Itanium still sells fairly well is that it's single threaded performance actually is pretty good. And it will, IMHO, get better with Tukwila. And you still can't get it into your thick head that the one machine with the myriad of CPU cores is the T5440. It has bloody 32 cores with no less than 256 threads. Man that is half the threads of a maxed out M9000."

Again notice that I use CPU core, not CPU, not CHIP.. And I still stand by my point.

0
0

@Jesper frimann

Regarding T2+ got a speed bump from 1.4GHz to 1.6GHz - no one would qualify it as a next gen cpu.

As I said earlier when we had this discusssion back then: when Pentium 4 got a speed bump from 2GHz to 2.4GHz - no one claimed the faster cpu was next gen cpu from Intel.

Regarding POWER6+, I dont see it as the next gen. POWER6 and it's derivatives are one generation and should be pitted against Niagara T2 and T2+. POWER7 is the next gen and should be pitted against Niagara T3. The coming T4 next year, is next gen and should be pitted against POWER8 (not the current POWER7). If T4 is faster than POWER7 - so what? It is POWER8 the T4 must beat to claim superiority. It is obvious that newer generations will beat older generations - and nothing to brag about. I will not care or brag about if T4 beats POWER7 - it should happen. On the other hand, if POWER7 beats T4, it is impressive of POWER7 (or bad from Oracle to fail with T4). In any case, POWER7 is better than T4 if that happens. And Oracle must sharpen up. If it happens that next gen T4 is slower than current gen POWER7.

.

"...My claims are normally always backed up by 'facts'...." Oh yes? I dont agree. When I claim IBM Mainframes are dog slow, you vigourosly deny it - and later you confess they are not good at number crunching - i.e. has bad cpu performance. That is not a claim backed by "facts" from you.

Or when I claimed that Niagara T2 is 13x faster than IBM CELL, you said the Sun benchmark was "cherry picked" and what not. You did not accept the Sun benchmark - because it was from BMSEER.

When I quote BMSEER (which quotes other sources) I present FUD? In that case, I should in the future, never quote from BMSEER, but instead quote directly from the other sources that BMSEER quotes. (I dont understand how BMSEER quoting a white paper from IBM, makes that white paper to be FUD from Sun?).

When you present benchmarks where IBM is faster, I accept them. When I present benchmarks where Sun is faster, you deny them. Strange, eh?

.

You write:

"...The whole point is that it doesn't suck, power6 is damn fast.."

"..Again notice that I use CPU core, not CPU, not CHIP.. And I still stand by my point...."

Are you talking about cores or cpus?

You are mixing apples with oranges. First you talk about how fast the cores are (which they are), from that you can not infer that the entire cpu is fast. That is not logically correct to do. Ive explained that many times to you.

And then you suddenly claim that the cpu is fast. That is a logical contradiction. First you establish that one core is fast (this is correctly done) then you suddenly jump to the conclusion the entire cpu is fast (this is wrong to do).

Then you say: I talk about cores, not cpus. But in the first quote here, you talk about the power6 cpu. Not it's cores. Contradiction, again.

0
0
Grenade

I hate to wade into your personal war but...

I feel I have to make a few (succinct!) points as a performance architect who works for IBM (Hence AC but I've professed my 'bias')... But I get no commission for selling IBM stuff - I work across a number of processor/server architectures (Sun, Windows+Linux on Intel and IBM) and personally try to give as unbiased advice as possible - not that its relevant as you'll see.

Mainframe - I've created a number of performance models based on both benchmark and real data. Z is poor at running J2EE type apps that UNIX/Sun servers will perform considerably better at -- maybe the emulation needed in Unix System Services is to fault - I just work with the raw numbers I'm no Z expert. However anything that works nearer the bare metal of Z such as native DB2 performs incredibly well, raw IO performance is stellar. Not to mention the large number of specialised instructions such as those for crypto and hardware decimal floating point which will give large performance improvements for certain apps. Anyway people mostly buy Z for the reliability and security not the performance...

Performance of cores is mostly irrelevant. I am often involved in improving the performance of large systems - usually Unix and Java to be fair and I never find myself proposing people move to Power or Intel or Sun cores to improve performance. You can get orders of magnitude performance improvements through code optimisation and architectural changes which a change of processor architecture could never reach. In fact a very large online UK system - one of the highest peak online TPS - moved from Power 5 to Power 6 and saw a dramatic drop in performance (Gasp Kebbert is right!) but it was due to a poor choice of Java framework in the next version of the app made at the same time - some judicious optimisation/tuning got it back to 2x faster than before even with the poor code (So what did the core change actually do?). The roadmap is important to be sure - as is the scalability of the system (Lots of fast cores) especially where the app does not horizontally scale well.

I really could not give a toss about individual core performance for the most part -> this argument is mostly moot for organisations choosing systems. Its a pissing contest and a small part of the decision making process for a Sun/Oracle or IBM win.

Another small example for Kebbert - One large bank I work with have a number of new system Z machines and a lot of Power, HP, Sun Unix machines. Now the Z runs far more mission critical apps, processes the vast majority of the core bank critical online and batch transactions and costs considerably less than the 100's of Unix machines and the army of people required to run them. It has had 0% downtime in decades and unsurprisingly they are really happy with it - neither IBM with Power/AIX or Sunacle could displace it.... What a weird world eh - do they not know the cores are, like, really slow man!

(I'd point out the reason the nix estate is so over-inflated is due to poor use of virtualisation and very poor coding making for very inefficient code --> Both of which are rarely found on Z)

I realise my example are all IBM tech :-) well I am biased. But anyone running super fast Sunacle processor X or IBM processor W feeling really happy that they have the fastest system possible are missing the point and do not get performance -> Software is king! (Caveat: In 99% of cases - I'll agree in the HPC world things are very different -> but a lot of the real world run shitty inefficient (i.e. J2EE) apps) .

0
0
This topic is closed for new posts.