Posts by PowerMan@thinksis 88 publicly visible posts • joined 31 Jul 2013
SAP S&D Tier 2 benchmark delivers 481 Users per core on the SPARC T7 compared to 996 with IBM's POWER8. Memory bandwidth of 160 GB/s per socket with 4 memory controllers servicing 32 cores versus 230 GB/s per socket with 2 memory controllers servicing 12 x POWER8 cores. Speaks for itself.
After reading the article and the comments I'm dumbstruck by HP. They failed with Itanium, failed with the cloud, in software, shed their instrument business years ago and now splitting into two companies all while developing a brand new server, compute and memory technology running a new OS - on their own! Why for God's sake?
IBM's POWER8 servers using the CAPI http://www-304.ibm.com/webapp/set2/sas/f/capi/home.html interface attached to their FlashSystem 840 storage takes a 48 TB 2U Flash array and after using RAID with a Hot Spare they create a 40 TB RAID protect memory extender from the servers base 2 TB. Fast and resilient in 2015. The solution is called IBM Data Engine for NoSQL based on the Redis Labs NoSQL product. http://www-03.ibm.com/systems/power/solutions/bigdata-analytics/data-engine-nosql/
Power8 technology is roughly 2X greater performing over x86. Factor that in with hypervisor options of No hypervisor, PowerKVM or PowerVM then look at the cost of the Linux only servers and your assumption they are more expensive is no longer the case. Look at http://www-03.ibm.com/systems/power/hardware/linux.html to see for yourself. Remember these prices are list. Also note the performance, scalability and reliability capabilities which Power is known for and you can see Power8 has redefined itself.
IBM Power8 servers use CAPI or Coherent Accelerator Processor Interface shares a PCIe slot. It bypasses the device driver and PCIe stack placing adapters like GPU's, FPGA's, Graphics, Network and Fibre cards making it a peer to the Power8 core. Users can attach a IBM FlashSystem 840 with 48 TB of eMLC Flash NAND Flash chips on a Flash Module. Using their 2D & VSR Raid you have 40 TB of usable RAID protected memory. Now, you have enterprise flash that is reliable because of what it is made of, how it is configured with 2D and VSR Raid and serviceable because of the 840 hot swap features running at near memory speeds with direct access to all Power8 processors.
Info on CAPI http://www-304.ibm.com/support/customercare/sas/f/capi/home.html http://www-304.ibm.com/webapp/set2/sas/f/capi/CAPI_POWER8.pdf http://www-304.ibm.com/webapp/set2/sas/f/capi/CAPI_FlashWhitePaper.pdf
Info on Power8 http://www-03.ibm.com/systems/power/hardware/
Info on IBM Flashsystem 840 http://www.redbooks.ibm.com/abstracts/tips1158.html?Open#contents
First off I don't talk about SPARC's demise to customers - only on these forums in response to ongoing, incessant, nonsensical rants. They (customers) are the ones who tell me about it. I am happy to tell someone who will listen (hint hint) all of the things the Power platform does great and if asked contrast that to competitive platforms. I don't need to disparage competitive platforms to customers to make Power look better; it's capabilities speak for itself. I know you enjoy attacking and tearing Power apart because you fear it and it threatens what you know and like. Understandable.
Don't care about M6 other than Power8 E870 delivers almost 4X the number of SAP Users per core. That translates into less software licenses and cost while doing more work on each processor. Stronger cores make for more efficiency. Yes, that is like light to a vampire isn't it Count Dracula?
With regard to M7 - I'll take the bet now that M7 won't be twice as fast as Power8....not on a per core basis. Do you see the theme yet? Not moving off of it. Since SPARC can no longer make powerful computers they have to make big computers so they can aggregate the individual pieces to claim victory - pitiful really. If the M7 shows up, we will see what public benchmarks are released that have current competitive results for comparison and not Oracle internal benchmarks or obscure benchmarks where the last Power result was a Power5 595 from 2007.
On your last T2+ and Cell comment - it's time to take your medicine again. The shadows are coming out and beginning to attack you.
SPARC is dead. Oracle's goal is to sell crap whitebox x86 servers that cost them little while they layer tons of expensive software on top of it then stick a label on the outside called "ExaSomething". Oracle is a software company. It makes sense. I don't blame them. That is there identity. Sun and SPARC is a mindshare and reputation they can leverage to sell the other products. The traditional SPARC servers from Sun (i.e. UltraSPARC) as they began OEM'ing Fujitsu's variation of SPARC. Further, they bought Afara WebSystems to get the Niagara processor which is the genesis of the T5. So, if you look at what Oracle is peddling today, SPARC is dead. If you want to stand on the fact that a SPARC derivative exists running Solaris, fine. It's a decent chip for the T series. It finally doesn't suck after the first 3 generations did. It roughly has the per core performance of the P5/P6 servers and with the exception of having 4-8X as many cores as everybody else so you can have massive numbers it doesn't hold a candle to any competitive solution. There are benchmarks and every one can be dissected and dismembered with ease because Oracle uses games like "per incident support" and using 1 year licenses instead of perpetual licenses because lots of customers buy 1 year licenses don't they? No, I'll stick by my statement. Oracle is a cheater and liar. Hitch your horse to them if you like.
Dude, the SAP S&D 2 tier benchmark site is not a secret. Anybody can look at it. Why do I need to provide a link? Here - http://global.sap.com/solutions/benchmark/sd2tier.epx What you need to see is not a "Fast" result from Fujitsu or Oracle from "Aggregated" result for lots of weak technology. That is why Oracle is releasing a 1024 core M7. If they could make a killer 256 core M7 they would but they can't. All they can do is make lots of little Timex watch cores, glue them together, add up their individual performance numbers to get the number they desire so Larry can pound his chest saying "I will be like IBM". Wow, pathetic! 883 and 996 is all you need to know. That is the # of Users per core for Power8.
Do you see squirrels and shiny objects? That article uses Intel material that came from their Ivy Bridge announcement. Go figure they made those claims. It is also preposterous. Your extrapolations are wrong again and further damage your credibility.
The benchmarks are out there. Go look for yourself. SPEC, SAP, Oracle.
No, I said the Power Linux servers have price parity. This is a volume play by IBM for customers who want the reliability of Power servers which x86, SPARC and Itanium can't touch with a ten foot pole, hypervisor choice of PowerVM, PowerKVM or no hypervisor (yes, never thought I would say that) so they can drive the maximum out of the server while run SuSE or Ubuntu in Little Endian Mode, RedHat (and SuSE) in Big Endian today and LE soon as well as Debian in BE. Just compile x86 source and run it on Power.
Not sure how you compare the E880 (not P880) to a 8 socket 12 TB x86 server - I've done my best to not call you a Troll here but with idiotic statements like this it makes me think you are just a flame thrower, trouble make.
Larry and John say a lot of things. They are both full of sh*t - you can tell them the next time you lick their boots. Unless it is a audit able benchmark I don't believe anything they say.
The only one saying they are exiting hardware is you - broken record. Your also on repeat with the 15 Power6 procs vs four T2+ procs. Listen and listen well so it sinks in. The fourteen you reference is fourteen cores or 7 dual chip modules or 7 processor books or 7 sockets. Your T2+ with four procs is actually 4 x 16 cores or 64 cores. So, you are trying to convince the reader a SPARC T2 with 4 beats a Power6 with 14 when in fact it is SPARC T2 with 64 and Power6 with 14. This is what you do, why you have no credibility and why it's not worth wasting my time answering each of your other crazy points.
I have a open offer to Oracle and you @MadMike. We can meet at any major airport (easy to fly into) to have a publicly attended and recorded debate on these topics. You can bring whatever and whoever you like. We will each have a whiteboard and wifi but only you (the opposition) and I get to talk. We will invite a impartial panel to moderate and score. You can continue to make false and inaccurate statements on here with little accountability. I will stand before God and country with my beliefs. If I am wrong I'll say so.... but if I am right....well, I might smile and enjoy it.
All, I want to apologize to the average reader who has had to endure my rant against @MadMike (not to be confused with @Mad Mike). @MadMike as seen at this article http://forums.theregister.co.uk/forum/1/2014/09/29/ellison_sparc_m7/ and many, many, many more articles all over the web looks for articles on technology, usually SPARC and Power to not only bash Power technologies but flat out lie about it. He/She does so with no shame or regard for the impact it has on the reader nor your desire to obtain information or participate in a cordial discussion.
I worked at Sun for 10 years, IBM for 4 years, an instructor in the Army on primarily SunOS and Solaris and now work for a IT BP where I run our Power practice. I enjoy technology, desire competition and respect competitors. I don't suffer liars and cheaters very well. There are some vendors, sellers and fanboi's who will do and say anything to advance their cause. I answer to a higher power (different One this time) and strive to keep my scales balanced. Proverbs 20:23
http://image.slidesharecdn.com/power8-announce-140529054737-phpapp01/95/ibm-power8-announce-3-638.jpg?cb=1401360585 I am a IBM BP and Power Champion. I was just in the labs prior to the P8 announcement. I saw OpenPower, stuff coming in 2015 and a few other things I can't mention the names and not germane to this discussion. I didn't see any P9, not to say it doesn't exist but I was in the P8 labs. I did talk with the P9 architect though, mostly on P8 vs P7 topics. We've talked previously on P9 (to the extent I could pull info out of him). My point is these guys are not to share info in general (they just like talking to somebody with a pulse). It is the marketing people who decide what and when they will release info. I could contact Austin right now and ask for details on Power9 and I would hear things like "It has more cores, more cache, more this and that with these and those accelerators. We are targeting performance increases in the *X over P8"." That is what a Power Champion and the 2014 Power Sales Excellence Awardee would receive...unless I can get that P9 architect alone with a bottle of Double Oaked Woodford Reserve. That may not satisfy you (I hope it does somewhat) but short of the world ending tomorrow I expect to be selling Power for years to come.
@Capt'n - Businesses fail when they let emotions cloud sound decision making. If they need the output of the fabs but it cost more to run and upgrade then find a solution. They found someone who specializes in this. It costs IBM $1.5B per year. They are paying GF $1.5B, payable over 3 years for 10 year rights to the output. They will save $1B each over the next 3 years and not have to spend $1.5B for years 4 - 10. How is that not sound decision making? Frankly, we should credit them for making these kinds of common sense decisions - if it was your home budget we would say, stop the cable, sell the car, downsize. This amounts to a 10 year outsourcing deal that will give IBM the output they need while developing the next generation technology and fab source.
If you don't like the "per core" conversation how to you control (and reduce) software licensing/maintenance while delivering the consolidation and performance required for all workloads? You either say "per core" performance matters or it doesn't. The sum of the cores equal the chip/socket performance so it needs to matter somewhere. Intel has tapped out their architecture and is just getting the small performance yields from die shrinks and other tweaks while adding more cores - focusing more on mobile and reliability. Oracle is first and foremost a marketing company - hype and hot air.
the SPARC T5 & M6 chips are about having the max cores and threads to grab headlines. Everything else is an adulteration of the facts. Even if there is a speck of fact anywhere with the product it doesn't matter because they choose to make preposterous claims (WSJ ads, @MadMike, PhilDunn, YouKnowWho, YouTube, Brad Carlisle, Scott Lynn, John Fowler, Mark Hurd and King Larry himself). The body of work is readily available for anybody to find using their favorite search engine.
But, if you want to see the impact of "per core" performance then look at my blog where I compare Oracle on x86 vs Power then introduce DB2 as an alternative. http://powertheenterprise.wordpress.com/2014/07/25/tired-of-oracles-exhorbitant-pricing-try-ibms-db2-v10-5-on-power8/
Happy Troll? I wasn't going to respond to you but you've sucked me in because that is what you do with your crap! Let's look at some SAP S&D Tier 2 benchmarks and put your "Awesome" M10-4S into perspective shall we? The 640 core (wow, that's big - it must be the Worlds Fastest) delivers a whopping 239 Users per core along with 1319 SAPS per core. The "More Awesome" 128 core T5-8" is 312 Users per core and 1726 SAPS per core. The "Most Awesome" 384 core M6-32 and it's 1 of only 2 public benchmarks delivers 365 Users per core and 2067 SAPS per core. Wow - SPARC is AWESOME - Can't wait for the M7 which will be 4X.
Power8, Troll delivers this today: The 80 core E870 for the SAP S&D 2 tier is 996 Users per core or 5425 SAPS per core while the 24 core 2 socket S824 delivers 883 users per core and 4828 SAPS per core. because I can't trust you to get this right I'll spell it out for you Troll. The E870 is 4.1x greater than the M10-4S you cited. What do the you think the chances are the 192 core E880 will smoke the 1024 M10-4S? Yeah, pretty damn good. The 2 socket S824 beats the 8 socket T5-8 by 2.8x.
I can cite many, many examples Troll in many areas the games you and Oracle play to confuser readers. What you do is piss them off because they get frustrated by this kind of crap. Crap by my rants and crap by your lies. Because IBM calls a core a processor and Oracle calls a processor a chip or socket you will say "Our 8 processor beats IBM's 8 processor by 8X. Really? That is 8 cores vs 64 cores - sleep well at night Troll if that is how you have to sell your products. I just show customers how Power controls their software licensing costs. How it is the most secure platform on the planet short of the mainframe. Virtualization without limits.
And the comment about AIX not recertifying is definitely valid Troll. ISV's don't like to retest their software. By IBM releasing TL's to AIX 6.1 and 7.1 they reduce the rate of change on ISV's which means customers get the benefits of new platform enhancements without having to recertify the software stack. Since you are in Oracle marketing I can understand why these operational annoyances wouldn't mean anything to you since you haven't been in the trenches before.
Spewing talking points there Troll! No portfolio. What systems portfolio does Oracle have? How about HP? Cisco? EMC? Dell? Fujitsu? I would say IBM having the Mainframe with a substantial software and services business is pretty substantial. IBM software has some outstanding products from DB2, Cognos, WebSphere, Guardium, TSM, and a 1000 more - seem substantial to me. IBM storage has entry, mid-range and Enterprise storage; traditional and virtualized solutions. Their Flash technology was the #1 seller in 2013. Used as a standalone it can be paired with any server or any other vendors storage as a read accelerator. Used behind a SVC device, it's competing head to head with Pure and others that are full featured microsecond flash solutions. Power8 outperforms x86 by 2X, SPARC by 3-4X. The Linux on Power servers have price parity with x86, the traditional Scale-out servers are price comparable. These are the 2 socket 24 core servers that match Ivy Bridge 4 socket 60 core servers in performance. Customers that demand maximum reliability, serviceability, consolidation and performance will go with enterprise servers. These servers scale to 192 cores and 16 TB of Ram.
You keep talking about these 120GB SQL queries Troll. Oracle internal benchmarks I'm guessing?
So Troll, you said Rock was slow earlier yet it has features going into the M7. That was obvious from public collateral. Thx for linking the two though. Since Power9 would be a Tick, I would expect it for at least 3 years from the release of Power8. That is what a Tick-Tock is Troll.
Nothing SPARC has beats the sh*t out of Power. Not Power4, Power5, Power6, Power7 or Power8. Here is the scoop Troll. Sun and Oracle's approach is to create servers with weaka$$ cores but lots of them. 256 then 384 and 512. You run a benchmark to get a result. Because you have 512 cores and 32 chips you claim victory and the worlds fastest processor because that is what Troll's do. The reality is that your servers are pieces of sh*t and you have to lie about Power because that is who you fear the most. Power8 is delivering roughly 2.7 - 4X the performance per core on about every benchmark that are common between them. I'll add that Oracle doesn't publish on too many benchmarks - I have the list so please challenge me so I can make you look like a bigger idiot than you already are.
M7 is nothing more than a 32 core server on a chip with 8 clusters of 4 cores. The latency and NUMA affects between clusters and chips will be comical. Any benchmarks produced will be Oracle internal, synthetic, non-auditable or obscure ones so as to not be challenged. We know you want to take your 1024 cores to get a big result to claim the Worlds fastest processor. yes, congratulations, Oracle has produced 6 generations of SPARC servers in 4 years is the claim I think - wow, I'd be proud of that. Keep cranking out the junk.
Selling the fabs has nothing to do with weakness. Showing your cards there Capt'n, got an ax to grind? IBM's xSeries weren't weak from a technology standpoint, it's just that IBM can't sell volume. They are a high value IT innovator. Commodity is better for those with efficient supply chains like Dell and Lenovo. I watch my x86 counterparts in my company make a pittance selling x86 while the vendors make the money. Cisco, Dell, VCE walk away laughing while BP's make a point or two peddling their crap. Those kinds of margins impact the quality of the product, quality of the sale, quality of support - qualtiy of the solution, in other words you get what you pay for.
To say that Power performance is stagnant if not weakening is just ignorant. Power8 delivers 2X (in general) over every Haswell and Ivy Bridge processor out there. About every benchmark I can think of shows something like 1.7X - 2.3X per core greater performance with Power8. SAP 2 tier with E870 delivers 996 Users per core compared to 458 with Haswell.
I never heard anybody from IBM say SPARC64 is dead. I've heard many a customer, pundit and competitor to include IBM and partners alike say that SPARC is dead Troll. Once again you twist things. You always spew lie after lie. I could give you link after link for benchmarks and you would come back the next time and claim you never get any links Troll. Nice attempt at being the victim but failed again.
What kind of Roadmap do you want Capt'n? If you went to the recent Enterprise 2014 conference in Las Vegas (Oct 6 - 10th), IBM showed many charts with Power9 on them. Jeff Stuechelli also spoke on it. Power8 was just released. They typically have their version of "tick-tock" so the tock will come next as a Power8+...usually in the 18 - 24 month range. It's an average once all of the products roll out.
For the larger audience out there though, they do not need an NDA to see a Roadmap for Power. However, if somebody wants the details of unannounced products then every vendor in the world does require a NDA. 2 entirely distinct issues. When I was an IBMer I wouldn't know the details about a product to be announced until roughly 6 months beforehand. Details were scarce. Specifics are speeds & feeds weren't available until 1-2 weeks out. If I wanted more then I had to attend a NDA session with a customer with engineering to hear the same thing they did. No vendor wants details on unannounced products released so the competition can have FUD ready, change their release schedules in response, etc, etc. It's just common sense.
If you think the stupid map showing the chips means something, it doesn't, that is out there for Power9. It is also out there for SPARC, Itanium and Intel. Has Intel been accurate with their roadmap (x86)? Itanium? SPARC (as I laugh)? Power? I can answer for the Power, definitely yes although I will qualify it to say that they were late in shipping Power8. Less to do with the chip technology but more to do with the entire PureSystems distraction in the business that was just sold to Lenovo.
@MadMike You are a troll who looks for IBM Power articles to spew your constant lies and hyperbole. Your agenda and goal are obvious. You get off on stirring things up. In a courtroom you would be labeled a vexatious litigant or serial abuser of these forums. Responding to you is pointless as you lack character.
IBM is Oracles largest customers in terms of the platforms their software runs on. Also, far more Oracle runs on Power than DB2. So, something changed and happened from the P4 days. As a side note. I think Oracle should actually abandon (by selling Solaris & SPARC) to focus on it's software stack optimizing it to run great on every platform. Let it exploit the best features of Power, Intel, AMD, ARM, Itanium, MIPS, whatever runs it. They are slowly trying to lock themselves out by tying everything together.
Which is why IBM is investing $3B to cross that canyon. They are already approaching it with a tremendous performance and feature lead over x86. SPARC isn't even in the game. They are going wide and big (ie M7 with 32 cores per socket & 1024 cores per server) that *if* it ever comes to market and *if* it even produces a SAP S&D 2 tier benchmark will be comparable to the 192 core Power8 server. At this point it is about innovation which the people I work with at IBM in Power engineering have it in spades.
Intel has a better chip making technology because of "why"? Both Power8 and Haswell use 22nm. Power8 is delivering 2X the performance in SPECint/fp/jbb and SAP off the top of my head. I recall Power uses something like 15 layers in their manufacturing process compared to something like 7 with Intel. When you stack the highway you get 3D efficiency so if anything I'd say IBM's approach is the better technology even if their fab appears dated by mass production standards.
Can't tell if you are an instigator, competitor, maybe a disgruntled (ex)IBMer but because your boldness to claim what will happen with Power in the future I'll go with a competitor trying to inject FUD into the conversation. As a IBM BP who is invested in the product and who has a reason to be concerned - know that I am not concerned. I saw Sun use TI, UMC, TSMC. Guessing Oracle uses TSMC. I could look it up but don't care. Point is they use somebody and if you couldn't tell where Dell, HP and Lenovo get their Intel chips from, yet Lenovo and not from their own fabs. Good try IBM competitor.
Your comments started out as a sentimental note reflecting on memories of your youth only to have it turn into an attack piece full of unfounded statements. There is no reason to think that Power is dead because of this. It makes complete sense for IBM to continue using the technology, processes and personnel that built the chips today as they work on future products. They will begin working with Fabs - probably GF but maybe others like TSMC to see who can deliver the quality products needed at the price point.
Ah, there is the Oracle FUD Machine! "MadMike" who has hijacked the moniker of "Mad Mike". A Oracle marketeer not interested in contributing or even stating competitive facts rather simply stating LIES. Yes, LIES.
Does little good to refute him as he/she isn't interested in the facts as he/she has an agenda which is to spread more FUD about IBM, AIX and Power while trying to give the impression the only and best choice is Oracle, Solaris and SPARC. The ONLY thing he/she said that is true is that the E880 can have up to 16 sockets with 16 TB Ram. MadMike is pathetic.
Does anybody think a vBlock would be as appealing or successful as it is (to whatever degree that is) if it didn't use Cisco UCS? I contend UCS is what made the solution different from a Dell or HP (similar) solution. IBM's former Flex product doesn't matter at this point so I'll exclude it from further comments. EMC + VMware + HP ProLiant = boring. Same for Dell. Cisco completed the solution. VCE's services gave it the services wrapper with luxury feel. I personally think it was too expensive, over-hyped and inflexible solution but to each their own.
They had better think twice about building their own x86 servers. They need volume to offset the R&D. Lenovo has the Supply Chain, cheap labor , other influences. HP is splitting into two businesses separating the PC & Printing business from the Enterprise. Dell went private to get out from the pressure of wall street. IBM sells the x86 business to Lenovo. Cisco's x86 biz appears to be ok - not sure as Cisco itself is also showing some economic strain. Best to find a partner.
Twice as fast based on....what? The M6 has 2 benchmarks that I can find; SAP & EBS. Power8 E870/E880 has been announced for 15 days and has results for SPECint/fp (rate+base), SPECjbb2013, and SAP S&D 2-tier.
Let's be clear as you continually espouse this lie. SPARC is not faster than Power. SPARC has larger servers than Power which produce larger results. You then market that as being faster. Since most software, especially Oracle where you work licenses by the core it really doesn't matter if a server has 1024 cores rather if each core is the strongest it can be. This is why there are sporadic and few SPARC benchmarks. When there are results they are cherry picked (not a bad thing btw) and often in benchmarks that the competition doesn't use, recognize or "Oracle Internal". Further, Oracle marketing makes their usual overstatements and misstatements (i.e. lies) such as 4 T2+ processors beat 14 Power6 processors when The T2+ processor is a socket and the 14 Power6 processors are cores. Thus it is really 32 vs 14. Mistake? Possibly if there weren't many, many more examples like this.
For the public reader. The E870 is producing 996 Users & 5451 SAPS per core. A M6-32 delivers 365 Users & 2067 SAPS per core. Btw, the 640 core Fujitsu M10-4S is just 239 Users & 1319 SAPS per core. What did you say the M7 will be? How is the M6 faster than a Power7? Oh, I didn't show it did I? Here it is for a 780 at 594 Users & 3247 SAPS per core. Hmm, M6 isn't faster than Power7 and isn't even close enough to suck the exhaust of the Power8 machine as it is that far behind.
Solaris on SPARC is not bad. It's solid. If a customer wants it fine. However, if a customer wants the maximum performance, availability, reliability, flexibility, security, virtualization features that control software licenses costs with the lowest cost of ownership they will choose Power. It's not competitive, it's not inflammatory, it's not provocative, it's a fact. I am happy to have a public debate against Oracle to prove this all. Bring your best DE, PE, SE or whomever and get it on. We will record it for public airing and have a panel of independent panelists consisting of a customer, academia and analyst to arbitrate.
You are hopeless. Check out my two blogs at powertheenterprise.wordpress.com. I don't want to keep going back and forth as it bores the other readers not to mention me. For the record, I was kidding you about attacking me. I thought I made that clear in how I wrote it. Lastly, I encourage you to call the "Brett Murphy" who is at IBM. Look up his phone number at the IBM directory then call and leave him tons of vm. Others can find me at www.thinksis.com moonlighting from my day job :)
No, "you" and the misleading SPARC sellers like to talk about who has the highest score while not focusing on how many cores and other resources it takes to achieve that number. I think it is great to be in 2nd place with Power servers on various benchmarks as it almost always shows a Power server with 2X fewer cores with a better score for those cores than the competitive server with 2 - 4X more cores (implying lower performance for those cores).
Where are your links in this response? Check out my twitter "@PowerMan_SIS" and you will see I provide all kinds of quotes, pictures, charts, etc. These forum comment sections don't let us post anything but text which limits providing comprehensive sources.
I do not understand why you have to attack me personally. I am just trying to have a discussion with you while informing the other readers.........................I'm just messing with you! :)
Impostor "MadMike" pretending to be "Mad Mike" wants it both ways. 8, 16 and 32 sockets are the way to get the greatest performance. Clearly the *sum* is the strength and not the parts that make it up because he has a single socket T1- B with 16 cores when compared to another 16 core server the excuse is because it is a 1 socket and not two. A 32 socket server is the best not because it has 1024 power cores but because it has 32 sockets x 32 cores and its aggregated results. What "MadMike" doesn't want to admit but is obvious even if he won't acknowledge it (because the facts speak for themselves) is a SPARC server with a 16 co chip is inefficient. He says you have to compare one CPU to another CPU to get a fair comparison - seriously, all you mention are benchmark results for a T5-8 or a M6-32 with 8 and 32 sockets respectively then compare that to a S824 with 2 sockets or a Power7 780 with 8 sockets or fewer depending on the benchmark. This is one reason why sockets influence performance and why per core performance is the common denominator.
What you fail to grasp is that Power isn't always 2X, 5X, 10X (or more) faster than the competition. When talking about current generation processors, Power is not 10X greater than any of them. It is approximately 2X faster than Ivy Bridge and more over SPARC. It is however, 10X, 20X, even 50X more efficient. Thanks to the mainframe efficient hypervisor, how it dispatches and schedules workloads onto the resources it delivers performance and throughput. It's not about 8, 12, 16 or 32 cores per chip but the performance of each core. If those cores are located in different sockets, then the performance between sockets and so on. It's not the sum of all of those cores that is impressive but being able to allocate the cores needed for each workload.
You've lost it on your rant of Power6 vs T2. Not worth responding to. I can't speak for IBMers or the one you reference. Have them contact me and I'll set 'em straight. What I will say is this though. Even if Power servers (ie cores) were slower than the competition, because of the efficiency afforded by PowerVM which exploits the Power Hypervisor it delivers greater results with a better TCO. Your comment about "The core is faster, therefore the cpu is faster" is only relevant if you were comparing the same size chips - 8 vs 8 or 16 vs 16. But, it is absolutely fair to compare a 2 socket 8 core/socket vs a 1 socket 16 core/socket server. At that point the comparison is on cores and server.
Spewing more talking points and FUD. You aren't a "fanboi" as that implies loyalty to technology that may or may not be the best. You are dazed and confused...actually I would say you work for Oracle and simply defending the Larry's honor or you work in their marketing department. Maybe you are the new guy and this is part of your hazing - "Get our there and get your butt kicked while saying this and this and don't forget this!".
You can't be taken seriously when everything you say is an obvious attempt at disparaging competitive platforms with no substantial data. "Rumours say ..." - whatever!
This is a repeat of what I posted at http://www.enterprisetech.com/2014/08/13/oracle-cranks-cores-32-sparc-m7-chip/#comment-229928.
You made a lot of assumptions with nothing to back up any of your statements. Making the claim does not substantiate the claim.
Oracle achieves world records and impressive numbers by one means and one means ONLY. They produce servers with excessive sockets and cores then aggregate the values for the given product and claim superiority. That is weak engineering, disingenuous marketing with diluted value to the customer.
First proof point – Oracle sells software, primarily by core (yes, other means are available). Thus, it stands to reason that performance per core is crucial. Look at the SAP benchmark where there are Power7 795, Power8 S824 and the M6-32 results.
M6-32 with 32 sockets & 384 cores deliver 793,930 SAPS or 2,067 SAPS per core. P7 795 with 32 sockets and 256 cores deliver 688,630 SAPS or 2,690 SAPS per core. With these two, which delivers the greatest performance per core? The entry level Power8 S824 server with 2 sockets and 24 cores deliver 115,870 SAPS or 4,827 SAPS per core. For additional comparison purposes, here is the 8 socket 124 core T5-8 that deliver 220,950 SAPS or 1,726 SAPS per core.
Just about every SAP landscape that I see have a total SAPS requirement but but that isn’t a singular value. Also, the ability of the server to utilize the full SAP value of that platform is also critical – this is dependent on the ability of the hypervisor, server technology and OS to be able to drive utilization that is useful. What this means is the likelihood of a Power8 24 core server being able to deliver 115K SAPS is quite likely. The likelihood of a x86 server being able to deliver it’s full SAPS is not likely. Given the weak hypervisor used in M6 & T5 along with the less capable SPARC chipset which includes the less efficient CMT (threads) makes it less likely for these servers to be driven as high as Power. I’ll hold off on critiquing Domains and LDOMs pending your desire to engage in that discussion as I look forward to that.
Looking at the per core results of M6-32 I deliver more per core with the 4 year old P7 795 server and IBM’s entry level 2 socket 24 core S824 delivers 2.8X higher performance per core over the “Solves World Hunger” Oracle SPARC M5-8 and 2.3X over the M6-32 “King of the Hill” server. To put this in perspective, if we normalize the performance of the S824 with its 24 cores, that is equivalent to approximately 67 T5-8 SPARC cores or 56 x M6-32 SPARC cores.
If you are a SPARC customer then you have fallen for their smoke and mirrors marketing that performance increase comes by doubling the cores. That simply doubles your software costs. If you work for Oracle then I expect you to sing from the Larry song book :) What is impressive with IBM’s Power8 technology when you see that it is 2X the performance over Power7 and x86 is that they are doing that on a per core basis. Using the Oracle sizing method we claim we have 3X the performance over Power7 and x86 but that is because we are going from 8 cores in Power7 to 12 cores in Power8. That means nothing unless you are pay software licenses by the socket then per core and # of cores per socket is very important.
Hopefully you see the error of your ways. I am a IT business partner, Architect and evangelist on Power technologies. I also worked at Sun for 10 years where I was a cluster and storage specialist (also StorageACE) and an instructor in the Army on SunOS/Solaris servers. I say this to say that i am a fan of Solaris. I love and miss Sun – great people and culture. Plus, I have experience on the platform. I am not speaking here as a seller trying to persuade you or the reader that I’m right because of talking points. I continue to challenge Oracle to a live, face to face technical debate that we can record and publish for all to see and hear the discussions. We both can have a whiteboard to aid in our discussions. We should have a panel of independent industry experts who can award points based on winning each topic. Because we know there are some people (ie you) who will never accept defeat even if proven beyond the shadow of a doubt.
Here is the Oracle marketing FUD. 'Ol MadMike spewing BS. Let's show the readers right away how you lack credibility. IBM's 14 "cpus" as you call them are cores whereas the "four SPARC T2+ cpus" are sockets. Each socket has 8 cores for a grand total of 32 cores. See for yourself http://www.oracle.com/us/solutions/benchmark/apps-benchmark/siebel-8-14000-pspp-on-solaris-twp-166395.pdf
Where are some other T2+ benchmarks by the way? You are hanging on by a thread. Why did you jump into a Storage discussion to start discussing Sun SPARC vs IBM Power6? I could ask if you wake up feeling dirty, if you hurt animals as a child, if you live alone.....sadly, its obvious what the answer is. Get a life.
The social experiment of "good enough" technology running Windows or open source software is coming to a head with damn near every American impacted. From Neiman-Marcus to Nordstroms to K-Mart this covers about everyone. PF Changs to DQ. Organizations like Target chose "low cost" Dell servers running Windows in every store because of "cost". It was recently reported to have cost them over $250M while their business is down some 40%. When will these companies wake up and take (cyber) security seriously? If they did, they should start with AIX using PowerSC or Security & Compliance. I'm sitting at McCaren airport leaving the Enterprise (IBM Power & Z) conference where I just delivered several presentations (as a Business Partner) on PowerSC to customers. You have to protect the stack plus the perimeter. With built-in encrypted filesystems that use the encryption accelerations on Power servers, Trusted Execution, Security Expert plus PowerSC features like Real Time Compliance and the Trusted suite of Boot, Logging and Firewall. Time to get back to using secure technology for the right workloads.
You complain of FUD then respond with not only FUD but you also are flat out lying in your attempt to disparage Power technology. Criticize the technology but spell it out rather than broad and sweeping statements.You demonstrate an anger that goes beyond competitive differences. You cite TPM, here is a recent article from him on the performance of Power8 vs x86 and SPARC. http://www.itjungle.com/tfh/tfh092914-story02.html
Larry boasts on everything - good for him, I relate. He has $45B while I want $45B. I have to earn mine like he did. My analysis of his boasting is that it is mostly hot air. The most relevant silicon for SPARC was probably UltraSPARC as it transitioned SPARC and SuperSPARC for a decade dominating the industry with US I & II until the ecache issues and 9/11.
M7 seems more like an evolution of the T5 family. They want to give the appearance it is a server processors like UltraSPARC I-IV or SPARC64 but it's base is the T1000. It has improved greatly since then but everything about it shouts "excessive yet deficient". 32 cores sharing 64 MB of L3 is just 2 MB of L3 per core. Appears to be at least a 3 Hop if not 4 Hop design from a core in Socket A to a core in Socket B. That results in NUMA behavior and latency. 2 TB Ram per socket which is 64 GB per core vs 170 GB per core on a 2 socket S824 - that's important when heavy virtualization is used. 2-3X memory bw over prev gens - well they currently have approx 55 GB/s so at 3X that would be 165 GB/s for a single socket with 32 cores compared to 230 GB/s with 4 - 12 x Power8 cores. They don't say what a database accelerator engine another source show it as memory compression which IBM's Power7 has had since March 2010. It's hard to comment but another source shows it as having 8 offload engines. It's an interesting claim that memory compression which introduces cycles into the pipeline can also accelerate database queries. If somebody has more info on this and could share that would be useful.
I'm not that convinced x86 has a reliable and robust memory subsystem to support in-memory workloads. I'm sure that will annoy some as thinking they are grown up but think mission critical...maybe life critical in certain cases. Would you want that running on x86? If your Board or C-suite expects you to deliver a solution that doesn't break the bank yet will run 24x7 would your first choice be x86? I would ask why you wouldn't consider a Power8 server? If a 2 socket, 24 core S824 with 2 TB of Ram and using memory compression that could be 4.5 TB equals the performance of a 4 socket 60 core Ivy Bridge server at the same cost, would you not consider it? Compare the security, RAS and virtualization features vs x86, VMware and Linux to see for yourself. Yes, this sounds like a commercial but so was the article. Using the Ford vs Chevy analogy, I'm asking you Chevy owners to consider a Ford. Chevy isn't what makes your business a success. Rather the reliability, security and total cost of ownership does.
Looks like VCE's future may be in jeopardy as the new EVO solution has the V+E. Clearly Cisco's John Chambers claiming VMware is Enemy #1 was spot on. Will be interesting to read future articles over the next days, weeks and months on what EVO means to this triad relationship.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2023
