Slashdot Mirror
Oracle To Halve Core Count In Next Sparc Processor
angry tapir writes "Oracle will halve the number of cores in its next Sparc processor and instead improve its single-thread performance, a weak area for the chip but one that's important for running large databases and back-end applications. The next Sparc chip on Oracle's roadmap, the T4, will have eight cores on each chip, down from 16 in the current Sparc T3."
Oracle Ruins Everything
Not trying to be a smartass, but does it really even matter? Hasn't almost everyone already decided to move away from Sun/Oracle, excepting those with a tremendous investment in that area? Can their sales really do anything except go down on the hardware side? And reducing the number of cores can't help, as cores is now the buzzword, just like megahertz was back in the Pentium days. Even AMD had to fudge the model names back then to get people to buy the processors, which admittedly were faster per Mhz than Intel, but customers looked at raw numbers. I would think that cores would be the same, even with a more sophisticated buyer.
Tequila: It's not just for breakfast anymore!
I wasn't aware the Alpha was that bad. I thought it was simply that the benefit of the processors wasn't great enough to convince companies to move from the much cheaper x86 platform. I saw a couple of Alpha desktops and they were pretty impressive.
The world's burning. Moped Jesus spotted on I50. Details at 11.
is another veiled sun downsize?
Bill Gates was quoted as saying "320 cores should be enough for anybody".
Does it maybe mean more register windows?
Because that would certainly help things like Java, and presumably oracle.
Anybody know how often a large query spills registers?
Do daemons dream of electric sleep()?
I'm pretty sure this was on Suns roadmap. Higher throughput per thread. Higher clock speeds. So have Oracle deviated from the plan Sun had?
I don't think the author had any understanding of the history of SPARC or Oracle (Sun)'s product linup. Here is an informative interview from the useful Sun hardware oriented blog on the subject http://www.c0t0d0s0.org/ http://www.oracle.com/us/corporate/innovation/innovator-hetherington-191304.html
Plato seems wrong to me today
The reduction in cores from 16 to 8 was part of the Sparc road-map before Sun was acquired by Oracle. Despite a lot of speculation it appears Oracle is following through with the plans they bought from Sun.
Lurking at the bottom of the gravity well, getting old
ISTR benchmark after benchmark saying that they performed about as well as a Pentium Pro/II of the same clock speed, when running native code. Except they were doing 533 MHz when Pentium Pros were doing 200. Oh, and the benchmarks I remember showed that the Alpha could emulate x86 code as fast as the Pentium Pro 200 could run it natively, after DEC's emulation software had profiled the code.
The problem is this... they were also, IIRC, more EXPENSIVE than said Pentium Pro machines, and they could (for the Windows market) only run NT, when everyone targeted 95. And the performance advantage was completely wasted if your code wasn't written for Alpha. (So, you could run Office 95 and such on them, but because Microsoft only compiled the OS and maybe some server software, for general desktop AND workstation duty if your business needed Windows, a PPro box was cheaper and may have been able to do the same job.)
(Keep in mind that back then, Microsoft was ambivalent about x86, at least in the workstation and server market. Windows NT was written to run on quite a few popular processor families - MIPS, PPC, and Alpha, in addition to x86. And, Microsoft made what was essentially an AT Architecture MIPS system specification for running NT on MIPS.)
um no...
Alpha systems were well worth 4x their other contemporaries.
They made things possible that were simply unable to scale on Solaris kit.
This is why they were also thrown into some of the early computing clusters and render farms.
OTOH: SGI are the poster boys for overpriced gear with lack luster performance.
A Pirate and a Puritan look the same on a balance sheet.
Reducing the core count lets Oracle make each core bigger, to add features making each faster. But can't Oracle keep the same core count, and instead of increasing the core count in the next generation the way most other CPU makers will, just add circuits to each existing core? Is it really necessary to reduce the count? Process size will probably also be shrinking in that generation, and new tricks developed, as usual. Can't Oracle just make a bigger chip, and also keep the benefits of the high core count Sun already achieved?
--
make install -not war
useless on the desktop
I had Alpha's on the desktop in '97. They were excellent and you are the very first I've ever heard characterize those machines as 'useless.' That's just silly. People still trade those things on E-bay and they get good prices. I watched tens of millions of dollars of revenue engineered on those machines.
It really depended on what you wanted to do. Sparc machines where great at IO and memory access. Alphas just had the shear grunt to do work (and yes they were running at over 1GHz when most processors where running at half that)
. SGI were crap but if you wanted to visualise it they could not be beat (hudge amounts of custom graphics hardware).
they were RELATIVELY useless.
i'm a fan of the MIPS architecture and loath things like hyper-threading, but throughput doesn't lie. there was obviously a lot of bad programming on the OS layer as doing anything in the file browser would cause the window to reload every single element and redraw it... super lag... hard to blame the chip for stuff like that, but there simply wasn't enough software optimized for the architecture.
but that doesnt really matter now does it? We know your application only supports 2 and scalability isn't an option.
the SGIs also outperformed the alphas for server daemons and we switched the math and computer science department web servers off the alpha.
that is when i first realized the Mhtz race was useless marketeering... what good is horsepower without proper gearing? the alphas seemed to lag behind everything else i had access to despite having almost triple the clock speed.
my first hand experience as funded by the national science foundation says otherwise.
Well let's see some numbers then. My firsthand experience (admittedly not funded by NSF!!!) says you're dead wrong. Secondly, who exactly claimed that Alphas were cheap? They WERE more expensive, but also more powerful.
do you have any insight of your own to provide? why are we all not using alpha chips if they were so affordable and powerful?
Perhaps you've heard the term "Wintel" before?
That's rather my point. The Alpha was anything but a bad chip, it's just that on a cost-benefit scale for what they were being marketed for, it made no sense. The problem, at the server and high-end workstation end, was that while the Alpha could certainly outperform Pentiums, the price made them very unattractive compared to the Intel was throwing out there.
The world's burning. Moped Jesus spotted on I50. Details at 11.
The DEC Alpha was (and still is) a brilliant architecture. The designers took great care from the start to make sure that it would scale, both in clock and core count. It was simple, elegant and fast.
IIRC, the early chips were fast enough to emulate x86 code at a reasonable speed. If all you wanted to do was run emulated x86 code though, then maybe they were "useless". This was especially true before the BWX extension, which introduced a number of byte oriented instructions.
Native code, on the other hand, would leave you with no misconceptions about the speed of the Alpha; it was truly impressive. DECs compilers and math libraries were also excellent.
IIRC the original athlon actually licensed some of the features of the alpha from DEC too...
When will people realize that not everything runs better on more cores, especially stuff that's highly dynamic say like a database query which is effectively a long sequence of conditionals. You talk to people and the first thing they ask is "yeah, but how many cores does it have"... it's like multithreading didn't exist until dualcore cpus.
A cpu has a limited amount of processing power; some things you can only do in sequence ergo you can't do them in parallel ergo you're limited by the core-speed ergo you're fucked with 16 core 1GHz machine against a 1 core 2GHz machine.
I have had various experiences. For some of my work the Alphas trounced everything. They were very fast processors. Larger data sets and I found that sunOS on fujisu (sparc) machines worked beter. Mind you I may be biased as the 16 proc machines I had access to were not quie comparable to the alphas (I think the alphas still out performed in floating point though).
But If you wanted to see your data then you had to have something from SGI. SGI really had impressive 3D hardware. Most low end 3D grphics cards can propbably out do SGI now but having to have 8 or 16 full length cards at a few grand a piece was fun. That was the first time I used 3D graphics.
It really depended on what you were doing too. Alpha's were built for raw speed and were good for certain tasks. Company I worked for back then used them for Lightwave rendering circa 1996/1997. But I believe the average box was somewhere in the neighborhood of $35,000 - $50,000 a pop for Dual 500 Mhz, 2GB of RAM and pretty much were all render nodes. Most of the workstations were SGI/IRIX and in the early days there were even quite a few Amiga around.
"The problem with socialism is eventually you run out of other people's money" - Thatcher.
At the very least, Oracle has introduced a great deal of uncertainty into Sun products, so you have to ask "What does Sun hardware offer than other hardware doesn't?". With all the bad press, they have an uphill battle converting people to Sun from other platforms, and for those who have a choice, what *exactly* is the big benefit that can't be purchased from someone else for less?
Do you care about crypto at all? If so, the T-series CPUs have on-die MD5, SHA-1, SHA-2 family, DES, 3DES, AES (multiple modes of operation), RC4, RSA (up to 2Kb), and ECC acceleration, as well as RNG. The T3s can do almost 80 Kop/sec for RSA 1024. All you have to do is link against the Solaris-provided OpenSSL library and call the appropriate "engine" APIs to activate things (this is built-in to a lot of FLOSS software already (e.g., Apache)).
The T5220 (T2 processor, the T3 just came out) has been benchmarked as doing 44 Gb/s AES128: and that's on the crypto co-processors, so the "real" processors are free to do "actual" work--like serving HTTP requests. At the same time as this, the T2 can also do 38 Kop/sec of RSA 1024. At the time this benchmark was published, a quad-core Xeon 3 GHz could do about 8 Gb/s AES1028 and 9 Kop/s of RSA1024 signing--with little to nothing left over to do anything else.
So you ask, "what can these systems do?" Well, how about: instead of paying for a bunch layer of load balancers to do SSL and RSA, and a whole bunch more machines to do actual web requests, why not just buy a lot fewer T2s (now T3s), and save power, cooling, and rack space?
The T-series is not good at everything, but for the mutl-threaded, multi-client workloads it was designed for it works very well.
the available native code was relatively non-existent.
alphas weren't much more expensive... if i remember right $800 got a 1.2 Ghtz bare bones system... about the same as a 433Mhtz with pentiums.
got it... so the ignorant public just never UNDERSTOOD how good the alpha product was and that is the reason they didn't buy it.
you're an idiot.
if only microsoft had dedicated itself to a symbiotic relationship with alpha to cross optimize... wahhhhh wahhhhh.
perhaps one day you'll not be an idiot?
Apparently, gratuitous insulting has become the cornerstone of logical arguments. Bravo.
Because Intel sued them over patents and buried the tech?
In case you haven't noticed MichealKristopeit* is a notorious sockpupeting troll. The amazing thing here is that he's somewhat on topic for so long.
Yeah I had a 486-DX100 and a 233 MHz Alpha 21064, both running Red Hat Linux.
The Alpha was so much faster for native compiled stuff, but I couldn't get Netscape for Alpha, and running Netscape for x86 under the EM86! emulator was as slow as browsing the web with a Python based browser at the time. They were both too slow to keep up with "fast" downloads like a 28k modem... So I wound up using the 486 machine as my graphics console, and running all of my batch stuff on the Alpha, with them sitting next to each other connected by cheap coax ethernet.
What amazes me is that I now have a quad core 2.4GHz Intel i7 Xeon with 12 GB of triple-channel RAM and gigabit connection to the internet here at work in a university, and I still get uncomfortable lags with browsing. Compared to my 486DX-100 and 20 MB of RAM, I am not sure I see that much more value in todays web to warrant this level of resource overhead. I expected us to be in the sci-fi future by the time we had this kind of equipment...
"That was the first time I used 3D graphics."
Perhaps I should say this was the first time I had used 3D graphics in anger. No textures...just trying to push polygons at the screen. trying to render voxels using GLUT/GL. That is when I started to apprecieate the SGI machine I had access to. It was too slow to compute the scene but nothing else could display it fast enough to get an idea what was going on.
That is what compilers are for. It is silly to suggest that there was no software for a Unix based system, especially at that time.
Availability of Windows based software would have helped, but was hardly necessary. The market didn't kill the Alpha, it was pure stupidity on the part of management.
Oh and this was using the shuttered glasses (LCD) that give you a headacke after ten mins. They claimed 30 hz but that did not really work. The pressure ball controll did work well.
alphas weren't much more expensive... if i remember right $800 got a 1.2 Ghtz bare bones system... about the same as a 433Mhtz with pentiums.
got it... so the ignorant public just never UNDERSTOOD how good the alpha product was and that is the reason they didn't buy it.
So what you're saying is that no, you don't have any performance numbers that can remotely back up anything you've claimed?
you're an idiot.
if only microsoft had dedicated itself to a symbiotic relationship with alpha to cross optimize... wahhhhh wahhhhh.
perhaps one day you'll not be an idiot?
Classy.
"Wintel" won as a platform because it was common, cheap, fast (enough) and yes, because Microsoft was very important to the PC landscape! Look at other chips like PowerPCs and so on that had some great performance and energy qualities but were dominated by Intel chips and Microsoft OS/software.
Is it true that you're a notorious sockpuppeting troll? I don't think I've seen your name before.
no, actually the norm is tongue in retarded cheek backhanded rhetorical implications like "Perhaps you've heard the term "Wintel" before?".
Yes, one can easily get into a pissing match over rhetorical techniques, like your "Funded by the NSF!!!" call to authority. Big whoop. My point was perfectly valid, and your inability to respond to anything I said lends credence to the AC who claimed you were a sockpuppeter.
slashdot = stagnated
Says the guy with the high 7-digit UID?
If you think the web is a bloated mess now, just wait until you see the hardware the sci-fi-future web will bring to it's knees!
From hell's heart I fstab at /dev/hdc
The Athlon used the EV6 bus from the Alpha, in fact the AMD 751 and 761 northbridge chipsets designed for the Athlon were used by Samsung for Alpha 21264 based systems!
There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
you're an idiot.
i've used all the systems first hand... the alphas ran the worst... obviously because of non-optimized software on every layer of the application stack. your inability to understand that the "AC" may very well be yourself as NO ONE has taken responsibility for the claims is hypocritically ignorant.
this account is NEW. you're the same old idiot.
bringing up such trivially irrelevant topics is again hypocritically ignorant.
would you rather i post with my 5 digit UUACCOUNTUSERID? you're an idiot.
if ANY user account CAN be a "sockpuppet" then ALL user accounts are "sockpuppets".
cower some more, feeb.
you're completely pathetic.
Depends upon the particular model. The DEC Multia was particularly problematic as it tended to suffer rather quickly from heat related problems and god help you if you were foolish enough to let it sit there flat while in operation.
would you rather i post with my 5 digit UUACCOUNTUSERID?
Yes, I would like that, please.
Revisionist history much? DEC sued Intel. This was...unwise, and Intel sued them back. The companies did, however, settle.
I said "Can HAVE cores plz"!!!!
"There is more worth loving than we have strength to love." - Brian Jay Stanley
What amazes me is that I now have a quad core 2.4GHz Intel i7 Xeon with 12 GB of triple-channel RAM and gigabit connection to the internet here at work in a university, and I still get uncomfortable lags with browsing.
I expected us to be in the sci-fi future by the time we had this kind of equipment...
One reason why so many things are still slow is because a lot of timeouts and delays are set to human-scale: on the order of seconds. Not milliseconds or microseconds.
The next reason is the speed of light isn't that fast.
And the last reason is what Intel giveth the Programmers taketh away. I'm guilty of this since I used to do 6502 machine code, but now write stuff in Perl (which on a modern machine can still do loops faster than a 1MHz 6502, but you can see where some of the speed gains have gone - I'm not sure how much work it would take for me to do a regexp match on a 6502 but I'm not going to even bother ;) ).
I was really hoping you would post with your 5-digit number :-(
did your mother name you "Moridineas".
you're completely pathetic.
Speaking as someone with an Alpha Station conveniently as a footrest...
I'm calling a double bullshit.
First of all the parent: 1.2GHz for 800 when Pentiums were 433MHz?
21164 Alphas, topped out at mostly 600MHz, and were faster than x86 clock for clock. They were also pushing the MHz limit at the time, being the first to 500MHz, and using a slightly later 533MHz (21164PC) they were damned fast compared to any x86 chip. As far as speed, at the time, they were overall the undisputed fastest processors. (DEC only made up to 8 processor machines as I recall. At least I think that's the limit for the 8400, a quick googling shows a reference to 12 processors, so it's likely higher)
21264 Alphas, were mostly released at under 800MHz, while air-cooled chips were demonstrated up to 1GHz, they weren't available. They were a lot faster per clock. (Of note, the Athlon used the EV6 (21264) bus which was one of the big reasons it did well. Plus a lot of DEC engineers going to AMD following the DEC/Compaq merger, which was if I recall correctly just before the 21264 was released, I don't recall any actual DEC hardware) Near the end P4s were released so you may be thinking 466MHz Alphas vs 1.2GHz P4s. (Where the Alphas would lose on integer, they would beat the P4s on floating point instructions) Benchmarking them, they were about 1.5x as fast clock for clock as at the time the fastest x86 processor (The original Athlon, mind you this is comparing optimized builds and the alpha didn't have SIMD instructions) If you were thinking of the API builds, you might be right, if you switch the MHz numbers around.
21364 Alphas were released following the HP merger and were the only ones which were over 1GHz. At this time, it was way late. The FPU was still respectable, but overall it was a case of too little too late. The 364 was not even a new core, it was a 264 wrapped in a much better communication to the outside world, on-board memory controller, and a very hypertransport like connection. (Which predated Hypertransport, at least in the beginning of the design, but was delayed so much, that as I recall Hammer wasn't that far off)
21464 was canceled, though it had a number of things, including the first Alpha SIMD instructions.
The alpha, even the 21064s, which were petering out in favor of 21164s when I got introduced to Alpha, were not cheap. However, they (21164PCs anyway) were priced comparably to a high end x86 system. When x86 got better and cheaper, they simply didn't keep up. Part of that was due to a DEC-Intel suit/resolution, which as far as I'm aware Intel didn't hold up. Which eventually got DEC merged with Compaq. Then it was all the 'Itanium is the future', where Compaq ended up basically killing Alpha for. When HP got it, they were also heavily in the Itanium camp. HP also had their own prior processor (PARISC) which was being killed off for Itanium. Which as I'm sure you are aware the Itanic future didn't turn out the be the case.
Second:
Slow, was the one thing the Alpha was not. Expensive, rare, hot and some other things, but slow it was not. Considering that an Alpha led the SPEC cpu benchmarks for over 9 *continuous* years, being broken of the integer by P4 on release, and fpu by a much later P4 (See above for how the Alpha's frequency wasn't keeping up on the 264s)
At no time, when alpha was being sold by DEC or Compaq, would SGI (MIPS) and Sun (SPARC) hardware have been faster per processor than Alpha. Look on the SpecCPU pages, or if you can find them (I can't) the old Seti@home statistics. Hell, look at just about *any* benchmark from around that time.
The DEC Alpha EV6 bus was licensed for the original Athlon (and continued with the Athlon XP). EV6 in the original Athlon form was a 100MHz double-pumped bus (200MHz effective) and very good in its day, far better than Intel's 100-133MHz bus at the time. In fact it was a major contributor to the Athlon's long-term dominance over comparable Pentium's for some time. In the end, it reached 200MHz double-pumped (400MHz effective) speeds. Intel didn't match that until Netburst, and we all know how well that turned out...
I don't recall any CPU-specific features were licensed from DEC... but I could be wrong.
did your mother name you "Moridineas".
Interesting question, but no, she didn't. It's actually kind of an amalgamation of a couple different things.
ah perhaps then we should use OS/2 instead of Solaris when they half single core CPUs..
you're completely pathetic.
you cower behind a chosen pseudonym. what are you afraid of?
Oh, could be any number of things...better safe than sorry! On the other hands, pseudonyms have a long and respectable history! You wouldn't criticize Publius, would you?
you're an idiot.
Hey, hey, hey, don't assign stagnated to slashdot! You mean slashdot == stagnated. (Unless you're writing Ada, then you're right)
Stop the brainwash
you're an idiot.
I see we have the same footrest / foot warmer. Mine's a lousy PWS 433, but it used to run loops around any PPro in its days. Still keep it for that little Quake game now and then... incredibly smooth gameplay...
Sun already tried this and had to give up. Sun's CPU design team just couldn't get the mode complex cored to work at competitive speeds.
I think that Oracle is making a big mistake here. They should let Fujitsu develop the big cores (which they are supremely good at) and themselves concentrate on the highly-multithreaded ones, which they are good at.
I don't see why this is a problem for developers, since the compilers take care of the CPU-specific details, and if your code is written in Java, it's a non-issue anyway.
Stick Men
Alphas ran VMS, NT, or Digital UNIX (nee OSF/1) none of which were designed ground-up for the user experience. But I had an Alphastation 3000/300X and it was tolerably peppy, so I think you're just imagining things.
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
The alpha was never intended to run NT really, but Microsoft was reaching out in all directions trying to see what would fly. Remember Microsoft Talisman? Neither does anybody else, but the hardware was essentially done when they canned it because they had totally misjudged the market. It was intended to be available as a big package which also included ISDN which was left in the dust by DSL and cable modems.
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
i remember in 1997 alpha processors ran over 1Ghtz when no other processors did, but they were useless on the desktop.
the problem is obviously attention to detail... engineers who do not take pride in the functional optimization of their products.
Drugs were better back then. 1997 Alpha processers were at about 4-500MHz. The 1GHz Alphas didn't come into 2002 or so.
My job is to maintain and enhance an industrial scheduling system. The G2-based platform is single threaded, as is the simulation engine (written in C++). It's grown from 3 concurrent users in 1999 to 31 today, and from 50 simulations per day to nearly 300. We've moved from SPARC to SPARC when new chips offered better prformance. The engine was written for, and so far can only be compiled with, Sun's compiler.
I'm sure we're not alone in having significant investment in single-threaded computational software. We've decided to ditch Sun, and a project is currently under way to port the engine.
It's a shame. I really quite like the stability we get on SPARC. But they've been too stagnant, and we've grown tired of waiting. Throw in the risk of losing support for hardware running a critical system, and there's no way we can stay on Sun.
Ah well, I was hoping you would get down-modded some more to save people in the future having to read your comments. Maybe next time! Cheers
Who started the suit doesn't much matter. Intel sued and won. That's why the Alpha is no longer its own line of chips. Any technology from the Alpha that Intel got in the settlement went into their chips. Any DEC got to keep went to Compaq then HP, and likely ended up in Itanium through them.
ur mum's face're imagining things
you do not believe individuals have the right to speak and be heard?
why do you cower behind a chosen pseudonym? what are you afraid of?
you're completely pathetic.
wow, forgot your meds again huh?
forgot your education? never received one? unable to learn, huh?
why do you cower? what are you afraid of?
you're completely pathetic.
Well, I didn't think I personally could silence you--you've shown more than enough willingness to keep talking! I did think that others might moderate you down (and I guess they did).
Publius!
Cheers
You show signs of serious mental illness, "Michael Kristopeit". You say you're a gun collector? Perhaps your neighbors should know...
If you want news from today, you have to come back tomorrow.
Since you're new to slashdot, here's the moderation FAQ, FYI: http://slashdot.org/faq/com-mod.shtml
you're an idiot.
Oh ok, sorry I couldn't help!
Except they were doing 533 MHz when Pentium Pros were doing 200.
Yep, when the so-called megahertz myth was true. P4 was Intel's attempt to abuse it.