| |
| 07 May 2008 19:48:31 |
| John Larkin |
| a dozen cpu's on a chip |
http://www.eetimes.com/news/latest/showArticle.jhtml;jsessionid=CESEXHWXUFJNKQSNDLSCKHA?articleID=207600531 I bet we'll see 256 one of these days. John |
| 07 May 2008 19:56:44 |
| Re: a dozen cpu's on a chip |
On May 7, 7:48 pm, John Larkin <jjlar...@highNOTlandTHIStechnologyPART.com > wrote: > http://www.eetimes.com/news/latest/showArticle.jhtml;jsessionid=CESEX... > > I bet we'll see 256 one of these days. > > John What is the current status of Microsquish regarding supporting more than two CPUs? |
| 07 May 2008 19:59:28 |
| David L. Jones |
| Re: a dozen cpu's on a chip |
On May 8, 12:48 pm, John Larkin <jjlar...@highNOTlandTHIStechnologyPART.com > wrote: > http://www.eetimes.com/news/latest/showArticle.jhtml;jsessionid=CESEX... > > I bet we'll see 256 one of these days. > > John I love the "nearly monolithic levels of performance" quote. Next we'll have CPU performance of biblical proportions! Dave. |
| 08 May 2008 07:42:04 |
| MooseFET |
| Re: a dozen cpu's on a chip |
On May 7, 7:48 pm, John Larkin <jjlar...@highNOTlandTHIStechnologyPART.com > wrote: > http://www.eetimes.com/news/latest/showArticle.jhtml;jsessionid=CESEX... > > I bet we'll see 256 one of these days. When you get to large numbers of CPUs it seems to make sense to stop making them identical. For servers this would be doubly so. Many of the CPUs won't need to do floating point operations. It also would make sense to do things like memory moves in the "Memory Mismanagement Unit" since the values don't need to be modified on the way through. This will make it a lot harder to say how many CPUs are in a chip. If there is only as much hardware as 200 full CPUs but 500 threads can be running at the same time, do you call it 200 or 500 CPUs. |
| 08 May 2008 07:43:29 |
| MooseFET |
| Re: a dozen cpu's on a chip |
On May 7, 7:56 pm, m...@sushi.com wrote: > On May 7, 7:48 pm, John Larkin > > <jjlar...@highNOTlandTHIStechnologyPART.com> wrote: > >http://www.eetimes.com/news/latest/showArticle.jhtml;jsessionid=CESEX... > > > I bet we'll see 256 one of these days. > > > John > > What is the current status of Microsquish regarding supporting more > than two CPUs? It depends a lot on what you call supporting. I think that NT could tie up two cores. |
| 08 May 2008 08:26:56 |
| Jeff Liebermann |
| Re: a dozen cpu's on a chip |
On Wed, 07 May 2008 19:48:31 -0700, John Larkin <jjlarkin@highNOTlandTHIStechnologyPART.com > wrote: >http://www.eetimes.com/news/latest/showArticle.jhtml;jsessionid=CESEXHWXUFJNKQSNDLSCKHA?articleID=207600531 >I bet we'll see 256 one of these days. >John Maybe. The problem is heat. The current E8500 Intel Wolfdale Core2Duo chips are about 110 sq mm for 2 cores and burn about 65 watts of heat running at 3GHz. Let's call it 50 sq mm and 30 watts per CPU. Extrapolate that to 256 CPU's, and we have 12,800 sq mm and 7.7kw of heat. If the chip were square, it would be 113mm on a size or roughly the size of a CDROM disk. 7.7kw of heat would make it the equivalent of about 15 coffee warmer hot plates running full blast. Using air cooling is out as it would require enough air flow to launch the PC into the air. Liquid metal cooling might work. At $0.15/kw-hr, this machine will cost $1.15/hr in electricity to operate (without energy management or shutting down un-used CPU's). Of course there will be improvements in technology, but using the existing available processes is a dubious proposition. -- Jeff Liebermann jeffl@cruzio.com 150 Felker St #D http://www.LearnByDestroying.com Santa Cruz CA 95060 http://802.11junk.com Skype: JeffLiebermann AE6KS 831-336-2558 |
| 08 May 2008 08:27:00 |
| John Larkin |
| Re: a dozen cpu's on a chip |
On Thu, 8 May 2008 07:42:04 -0700 (PDT), MooseFET <kensmith@rahul.net > wrote: >On May 7, 7:48 pm, John Larkin ><jjlar...@highNOTlandTHIStechnologyPART.com> wrote: >> http://www.eetimes.com/news/latest/showArticle.jhtml;jsessionid=CESEX... >> >> I bet we'll see 256 one of these days. > >When you get to large numbers of CPUs it seems to make sense to stop >making them identical. For servers this would be doubly so. Many of >the CPUs won't need to do floating point operations. Right. Amybe a few cpu's would have serious floating point power, or a few separate fp engines could be assigned to cpu's as needed. Lots of cpu's, doing stuff like file i/o or serial stuff, could be less powerful. I suppose we'll always need special graphics hardware, but just a few of those per chip. > >It also would make sense to do things like memory moves in the "Memory >Mismanagement Unit" since the values don't need to be modified on the >way through. > >This will make it a lot harder to say how many CPUs are in a chip. If >there is only as much hardware as 200 full CPUs but 500 threads can be >running at the same time, do you call it 200 or 500 CPUs. Next step is to get rid of task swapping and threads altogether. One CPU is the OS, and one cpu gets assigned per process. John |
| 08 May 2008 08:29:07 |
| John Larkin |
| Re: a dozen cpu's on a chip |
On Thu, 08 May 2008 08:26:56 -0700, Jeff Liebermann <jeffl@cruzio.com > wrote: >On Wed, 07 May 2008 19:48:31 -0700, John Larkin ><jjlarkin@highNOTlandTHIStechnologyPART.com> wrote: > >>http://www.eetimes.com/news/latest/showArticle.jhtml;jsessionid=CESEXHWXUFJNKQSNDLSCKHA?articleID=207600531 >>I bet we'll see 256 one of these days. >>John > >Maybe. The problem is heat. The current E8500 Intel Wolfdale >Core2Duo chips are about 110 sq mm for 2 cores and burn about 65 watts >of heat running at 3GHz. Let's call it 50 sq mm and 30 watts per CPU. >Extrapolate that to 256 CPU's, and we have 12,800 sq mm and 7.7kw of >heat. If the chip were square, it would be 113mm on a size or roughly >the size of a CDROM disk. 7.7kw of heat would make it the equivalent >of about 15 coffee warmer hot plates running full blast. Using air >cooling is out as it would require enough air flow to launch the PC >into the air. Liquid metal cooling might work. At $0.15/kw-hr, this >machine will cost $1.15/hr in electricity to operate (without energy >management or shutting down un-used CPU's). > >Of course there will be improvements in technology, but using the >existing available processes is a dubious proposition. The real advantage of multiple cpu's isn't performance; it's management. There's no reason to run all the cores flat-out all the time. John |
| 08 May 2008 08:34:21 |
| Jeff Liebermann |
| Re: a dozen cpu's on a chip |
On Wed, 7 May 2008 19:56:44 -0700 (PDT), miso@sushi.com wrote: >On May 7, 7:48 pm, John Larkin ><jjlar...@highNOTlandTHIStechnologyPART.com> wrote: >> http://www.eetimes.com/news/latest/showArticle.jhtml;jsessionid=CESEX... >> >> I bet we'll see 256 one of these days. >> John >What is the current status of Microsquish regarding supporting more >than two CPUs? Their policy is that for all improvements in hardware, there is an equal and opposite deterioration in software resulting in zero net gain in overall performance. In 1983, my 4.77Mhz PC took about 90 seconds to boot PCDOS 1.1 on a 160KB floppy. 25 years later, my 2.4GHz machine takes about 180 seconds to boot XP. My guess is it will follow the classic time management axiom. If a job takes one person 1 hour to accomplish, two people will take 2 hours, 3 will take 3 hours, 4 will take 4 hours, etc. Same with CPU's. -- Jeff Liebermann jeffl@cruzio.com 150 Felker St #D http://www.LearnByDestroying.com Santa Cruz CA 95060 http://802.11junk.com Skype: JeffLiebermann AE6KS 831-336-2558 |
| 08 May 2008 09:25:38 |
| Jeff Liebermann |
| Re: a dozen cpu's on a chip |
On Thu, 08 May 2008 08:29:07 -0700, John Larkin <jjlarkin@highNOTlandTHIStechnologyPART.com > wrote: >On Thu, 08 May 2008 08:26:56 -0700, Jeff Liebermann <jeffl@cruzio.com> >wrote: > >>On Wed, 07 May 2008 19:48:31 -0700, John Larkin >><jjlarkin@highNOTlandTHIStechnologyPART.com> wrote: >> >>>http://www.eetimes.com/news/latest/showArticle.jhtml;jsessionid=CESEXHWXUFJNKQSNDLSCKHA?articleID=207600531 >>>I bet we'll see 256 one of these days. >>>John >> >>Maybe. The problem is heat. The current E8500 Intel Wolfdale >>Core2Duo chips are about 110 sq mm for 2 cores and burn about 65 watts >>of heat running at 3GHz. Let's call it 50 sq mm and 30 watts per CPU. >>Extrapolate that to 256 CPU's, and we have 12,800 sq mm and 7.7kw of >>heat. If the chip were square, it would be 113mm on a size or roughly >>the size of a CDROM disk. 7.7kw of heat would make it the equivalent >>of about 15 coffee warmer hot plates running full blast. Using air >>cooling is out as it would require enough air flow to launch the PC >>into the air. Liquid metal cooling might work. At $0.15/kw-hr, this >>machine will cost $1.15/hr in electricity to operate (without energy >>management or shutting down un-used CPU's). >> >>Of course there will be improvements in technology, but using the >>existing available processes is a dubious proposition. >The real advantage of multiple cpu's isn't performance; it's >management. There's no reason to run all the cores flat-out all the >time. > >John Sure, if you have a general purpose application. However, if you're trying to solve a problem that requires massive simultaneous computation (examples previously listed in another thread), the chances are really high that all or most of the CPU's are going to be running flat out. Energy management is nice for reducing average consumption, but if you're grinding a weather system model, predicting the stock market, or run a brute force chess game solution, you want all that horsepower running all the CPU's at once. The leader in the CPU burning applications is currently the PC games industry. I've watched the "task manager" under Windoze to see how well Vista balances applications on a quad core HP machine. Actually it's the Process Explorer: <http://www.microsoft.com/technet/sysinternals/Security/ProcessExplorer.mspx Most apps use perhaps 10-20% of the available horsepower per CPU while waiting (yawnnnn) for the hard disk to finish doing whatever. Not so with the games. They use every CPU cycle they can grab to do predictive imaging (to paint the screen in RAM in anticipation of the next frame or move), variable image quality depending on available cycles, and such. 80% usage and up all the time is the norm. Little wonder game machines run hot. I got a better analogy. Suppose your automobile has 8 cylinders, but only enough radiator cooling capacity to keep 4 of them cool at a time. Think small water jacket and very small radiator. Around town, it will work as the engine energy management system shuts down cylinders that are allegedly not needed. For bursts of speed, you get all 8 cylinders. For idling, you only get perhaps 2 cylinders. For cruising, maybe 4 cylinders. That's fine until you decide to tow a trailer. Now, your average load is well over what the cooling system can handle and the engine melts. (There are engines that work on exactly this principle). Kinda like your energy managed CPU. The power is there when you need it, but only for a short time. (Comcast also calls it "burstable" download speeds). It works, but leaves something to be desired. Whether anyone will buy a machine that has the ability to rapidly run 256 simultaneous solutions, but only 1/256th of the time, is open to debate. Kinda reminds me of the early multitasking operating systems (Desqview, Topview, Windoze 2.0, etc). I would take a simple sieve benchmark and run it without the multitasker, and compare it with running multiple copies under the multitasker. I don't recall the exact numbers but the first Windoze, with cooperative multitasking, would take about 5 times longer to run as compared to running under DOS. Run multiple copies, and they would take far more than twice as long per copy (due to multitasking overhead). At the time, I was wondering if all this multitasking was really worth the huge performance hit. I guess it was. -- Jeff Liebermann jeffl@cruzio.com 150 Felker St #D http://www.LearnByDestroying.com Santa Cruz CA 95060 http://802.11junk.com Skype: JeffLiebermann AE6KS 831-336-2558 |
| 08 May 2008 09:24:40 |
| Joel Koltner |
| Re: a dozen cpu's on a chip |
<miso@sushi.com > wrote in message news:330b2006-558a-485b-8b82-9863651f6594@x19g2000prg.googlegroups.com... > What is the current status of Microsquish regarding supporting more > than two CPUs? It's all about licensing... Windows NT/2000/XP/Vista have always supported up to 32 CPUs at the OS level (it's 32 because somewhere there's a DWORD that has one bit per CPU for flags), and I believe that through NT 4 it actually would use however many CPUs it found. Starting with Win2K they started having difference licenses depending on the number of CPUs present -- the basic Win2K package would support 2. With XP it was clarified that the liensing would applying to the number of physical CPU *sockets* and not the number of *cores*, so while Win XP Professional with regular licensing still supports only two CPU sockets, if you find a dual-socket quad-core CPU, you can get 8 CPU cores all running at once. For the right price, Microsoft will enable more CPU sockets if you have them (they were targeting high-end servers), although even when four-socket motherboards were more common than they are today, the problem was often one of scalability: If you used the "regular" north bridge chipsets, performance didn't scale very well because the CPUs would get starved for memory bandwidth. Some companies built their own north bridge chipsets to get around this problem, but realistically the payback wasn't that great... which I think is now why it's atypical to see more than dual-socket motherboards for contemporary CPUs: People instead just get a couple of 1U-high servers and spend far less money for the same overall level of performance. ---Joel |
| 08 May 2008 09:31:28 |
| Joel Koltner |
| Re: a dozen cpu's on a chip |
Apologies for all the typos in there... I haven't had any caffeine this morning yet! |
| 08 May 2008 17:40:04 |
| Eeyore |
| Re: a dozen cpu's on a chip |
John Larkin wrote: > http://www.eetimes.com/news/latest/showArticle.jhtml;jsessionid=CESEXHWXUFJNKQSNDLSCKHA?articleID=207600531 > > I bet we'll see 256 one of these days. > > John But when will we see you cease using the greengrocers' apostrophe ? http://en.wikipedia.org/wiki/Apostrophe#Greengrocers.27_apostrophes Graham |
| 08 May 2008 10:34:39 |
| Paul Hovnanian P.E. |
| Re: a dozen cpu's on a chip |
miso@sushi.com wrote: > > On May 7, 7:48 pm, John Larkin > <jjlar...@highNOTlandTHIStechnologyPART.com> wrote: > > http://www.eetimes.com/news/latest/showArticle.jhtml;jsessionid=CESEX... > > > > I bet we'll see 256 one of these days. > > > > John > > What is the current status of Microsquish regarding supporting more > than two CPUs? That might be the minimum supported h/w configuration for their next OSA version. :-/ -- Paul Hovnanian paul@hovnanian.com ----------------------------------------------------------------------- Procrastinators: The leaders for tomorrow. |
| 08 May 2008 12:42:45 |
| Re: a dozen cpu's on a chip |
On May 8, 8:26 am, Jeff Liebermann <je...@cruzio.com > wrote: > On Wed, 07 May 2008 19:48:31 -0700, John Larkin > > <jjlar...@highNOTlandTHIStechnologyPART.com> wrote: > >http://www.eetimes.com/news/latest/showArticle.jhtml;jsessionid=CESEX... > >I bet we'll see 256 one of these days. > >John > > Maybe. The problem is heat. The current E8500 Intel Wolfdale > Core2Duo chips are about 110 sq mm for 2 cores and burn about 65 watts > of heat running at 3GHz. Let's call it 50 sq mm and 30 watts per CPU. > Extrapolate that to 256 CPU's, and we have 12,800 sq mm and 7.7kw of > heat. If the chip were square, it would be 113mm on a size or roughly > the size of a CDROM disk. 7.7kw of heat would make it the equivalent > of about 15 coffee warmer hot plates running full blast. Using air > cooling is out as it would require enough air flow to launch the PC > into the air. Liquid metal cooling might work. At $0.15/kw-hr, this > machine will cost $1.15/hr in electricity to operate (without energy > management or shutting down un-used CPU's). > > Of course there will be improvements in technology, but using the > existing available processes is a dubious proposition. > > -- > Jeff Liebermann je...@cruzio.com > 150 Felker St #D http://www.LearnByDestroying.com > Santa Cruz CA 95060http://802.11junk.com > Skype: JeffLiebermann AE6KS 831-336-2558 The processors would use finer geometry to some degree, which reduces heat since there is less charge to shuffle around. You can't make the chips arbitrarily big due to defect density issues. [Defect density will kill you way before heat is the problem. Just ask Gene Amdahl, or better yet the investors in Triology. Granted bipolar, with vertical current flow, is much more sensitive to defect density than CMOS, i.e. lateral current flow.].] The first dual CPU box I owned that actually worked as advertised is my AMD64 X2. Everything else had memory bandwidth issues. Incidentally, while Intel is a MCM, the AMD is monolithic. |
| 08 May 2008 20:42:42 |
| Nico Coesel |
| Re: a dozen cpu's on a chip |
John Larkin <jjlarkin@highNOTlandTHIStechnologyPART.com > wrote: > > >http://www.eetimes.com/news/latest/showArticle.jhtml;jsessionid=CESEXHWXUFJNKQSNDLSCKHA?articleID=207600531 > > >I bet we'll see 256 one of these days. I doubt it. A mailman won't deliver mail faster when he is using a Ferrari instead of bicycle. I think we are going to see slower, cheaper, more energy efficient computers with back-to-basic software for every day / office use. For example an embedded ARM or MIPS system running at 300MHz has enough power to run most common applications. It consumes at least 10 times less energy compared to a standard PC. -- Programmeren in Almere? E-mail naar nico@nctdevpuntnl (punt=.) |
| 08 May 2008 20:54:30 |
| Apostrophe Police |
| Re: a dozen cpu's on a chip |
On Thu, 08 May 2008 17:40:04 +0100, Eeyore wrote: > John Larkin wrote: > >> http://www.eetimes.com/news/latest/showArticle.jhtml;jsessionid=CESEXHWXUFJNKQSNDLSCKHA?articleID=207600531 >> >> I bet we'll see 256 one of these days. > > But when will we see you cease using the greengrocers' apostrophe ? > > http://en.wikipedia.org/wiki/Apostrophe#Greengrocers.27_apostrophes Where exactly in John's post did you see one of those? -- Rich Grise, Self-Appointed Chief, Apostrophe Police |
| 08 May 2008 22:26:45 |
| Eeyore |
| Re: a dozen cpu's on a chip |
Apostrophe Police wrote: > On Thu, 08 May 2008 17:40:04 +0100, Eeyore wrote: > > John Larkin wrote: > > > >> http://www.eetimes.com/news/latest/showArticle.jhtml;jsessionid=CESEXHWXUFJNKQSNDLSCKHA?articleID=207600531 > >> > >> I bet we'll see 256 one of these days. > > > > But when will we see you cease using the greengrocers' apostrophe ? > > > > http://en.wikipedia.org/wiki/Apostrophe#Greengrocers.27_apostrophes > > Where exactly in John's post did you see one of those? > > -- > Rich Grise, Self-Appointed Chief, > Apostrophe Police How about the subject line ? Graham |
| 08 May 2008 21:36:07 |
| Apostrophe Police |
| Re: a dozen cpu's on a chip |
On Thu, 08 May 2008 22:26:45 +0100, Eeyore wrote: > Apostrophe Police wrote: >> On Thu, 08 May 2008 17:40:04 +0100, Eeyore wrote: >> > John Larkin wrote: >> > >> >> http://www.eetimes.com/news/latest/showArticle.jhtml;jsessionid=CESEXHWXUFJNKQSNDLSCKHA?articleID=207600531 >> >> >> >> I bet we'll see 256 one of these days. >> > >> > But when will we see you cease using the greengrocers' apostrophe ? >> > >> > http://en.wikipedia.org/wiki/Apostrophe#Greengrocers.27_apostrophes >> >> Where exactly in John's post did you see one of those? > > How about the subject line ? Oh. Never mind. -- Rich Grise, Self-Appointed Chief, Apostrophe Police |
| 08 May 2008 17:47:06 |
| Spehro Pefhany |
| Re: a dozen cpu's on a chip |
On Thu, 08 May 2008 22:26:45 +0100, Eeyore <rabbitsfriendsandrelations@hotmail.com > wrote: > > >Apostrophe Police wrote: > >> On Thu, 08 May 2008 17:40:04 +0100, Eeyore wrote: >> > John Larkin wrote: >> > >> >> http://www.eetimes.com/news/latest/showArticle.jhtml;jsessionid=CESEXHWXUFJNKQSNDLSCKHA?articleID=207600531 >> >> >> >> I bet we'll see 256 one of these days. >> > >> > But when will we see you cease using the greengrocers' apostrophe ? >> > >> > http://en.wikipedia.org/wiki/Apostrophe#Greengrocers.27_apostrophes >> >> Where exactly in John's post did you see one of those? >> >> -- >> Rich Grise, Self-Appointed Chief, >> Apostrophe Police > >How about the subject line ? > >Graham > That's not one of them Greengrocers' apostrophe's-- rather that's a standard (now largely deprecated) way of pluralizing abbreviations. It's the method I was taught in school, and if you search the New York Times archives you'll find frequent examples up until a few years ago. http://en.wikipedia.org/wiki/Apostrophe "An apostrophe is used by some writers to form a plural for abbreviations, acronyms, and symbols where adding just s rather than 's may leave things ambiguous or inelegant. While British English formerly endorsed the use of such apostrophes after numbers and dates, this usage has now largely been superseded" Personally, I think it's less ambiguous in some cases so I'd rather see it used for all cases, but I guess the (spit) prescriptivists have won this round. Best regards, Spehro Pefhany -- "it's the network..." "The Journey is the reward" speff@interlog.com Info for manufacturers: http://www.trexon.com Embedded software/hardware/analog Info for designers: http://www.speff.com |
| 08 May 2008 23:01:24 |
| Eeyore |
| Re: a dozen cpu's on a chip |
Spehro Pefhany wrote: > Eeyore wrote: > >Apostrophe Police wrote: > >> Eeyore wrote: > >> > John Larkin wrote: > >> >> http://www.eetimes.com/news/latest/showArticle.jhtml;jsessionid=CESEXHWXUFJNKQSNDLSCKHA?articleID=207600531 > >> >> > >> >> I bet we'll see 256 one of these days. > >> > > >> > But when will we see you cease using the greengrocers' apostrophe ? > >> > > >> > http://en.wikipedia.org/wiki/Apostrophe#Greengrocers.27_apostrophes > >> > >> Where exactly in John's post did you see one of those? > >> > >> -- > >> Rich Grise, Self-Appointed Chief, > >> Apostrophe Police > > > >How about the subject line ? > > That's not one of them Greengrocers' apostrophe's Actually it is. See the article I linked to. A lovely example one sees even today in some British pubs on the menu is "fish and chip's". Maybe it should now be called the 'publicans' apostrophe' ? There's few greengrocers still around these days. > -- rather that's a > standard (now largely deprecated) way of pluralizing abbreviations. > It's the method I was taught in school, and if you search the New York > Times archives you'll find frequent examples up until a few years ago. > http://en.wikipedia.org/wiki/Apostrophe A damn ex-colonial abberation of course ! ;~) > "An apostrophe is used by some writers to form a plural for > abbreviations, acronyms, and symbols where adding just s rather than > 's may leave things ambiguous or inelegant. I've yet to see any written precedent for this. > While British English > formerly endorsed the use of such apostrophes after numbers and dates, > this usage has now largely been superseded" Quite so. > Personally, I think it's less ambiguous in some cases so I'd rather > see it used for all cases, but I guess the (spit) prescriptivists have > won this round. Eh ? Lets see; plurals should not be formed by the use of apostrophes. Seems simple enough to me. For example, for clarity, rather than pcb's I use PCBs. Would you seriously think of talking of diode's, resistor's or transistor's ? Graham |
| 08 May 2008 18:26:05 |
| John Larkin |
| Re: a dozen cpu's on a chip |
On Thu, 08 May 2008 20:42:42 GMT, nico@puntnl.niks (Nico Coesel) wrote: >John Larkin <jjlarkin@highNOTlandTHIStechnologyPART.com> wrote: > >> >> >>http://www.eetimes.com/news/latest/showArticle.jhtml;jsessionid=CESEXHWXUFJNKQSNDLSCKHA?articleID=207600531 >> >> >>I bet we'll see 256 one of these days. > >I doubt it. A mailman won't deliver mail faster when he is using a >Ferrari instead of bicycle. I think we are going to see slower, >cheaper, more energy efficient computers with back-to-basic software >for every day / office use. Maybe. But there's certainly no trend in that direction yet. Vista is even more bloated than XP. My conjecture is that multiple cpu's on one chip will make computers simpler and more efficient, and certainly more reliable. > >For example an embedded ARM or MIPS system running at 300MHz has >enough power to run most common applications. It consumes at least 10 >times less energy compared to a standard PC. Oh, things are coming along nicely: http://www.sun.com/emrkt/innercircle/newsletter/0407feature.html "Sun foresees the need for extending thread count beyond 64 separate instances of a computer, which is why the chip that will follow UltraSPARC T2 (currently code-named Victoria Falls) is being designed to have 128 threads. These threads, however, will be fully extendible. This will make it possible to link two instances of Victoria Falls — both sharing common memory through a hub chip — for a grand total of 256 threads." Just threads, not full cpu's [1], but it looks good. Here's one PPC and eight smaller processors on one chip: http://domino.research.ibm.com/comm/research_projects.nsf/pages/multicore.CellBE.html John [1] I kept the apostrophe so that people whose expertise stops at 20 KHz can have something clever to say. But to me, cpus looks awkward; you'd pronounce it "seepuss." |
| 08 May 2008 18:37:13 |
| MassiveProng@thebarattheendoft |
| Re: a dozen cpu's on a chip |
On Thu, 08 May 2008 18:26:05 -0700, John Larkin <jjlarkin@highNOTlandTHIStechnologyPART.com > wrote: > >Maybe. But there's certainly no trend in that direction yet. Vista is >even more bloated than XP. My conjecture is that multiple cpu's on one >chip will make computers simpler and more efficient, and certainly >more reliable. > Are you sure? Service Pack 3 was just released, and I would bet that XP is a lot more bloated now than it previously was. Funny, my Vista installation is still working flawlessly, and any error it ever did have was repaired either in session or on the next boot. I had one blue screen on boot, and it worked immediately after, and found the issue to be an Nvidia driver, which nvidia replaced with a newer, proper driver. |
| 08 May 2008 18:41:36 |
| FatBytestard |
| Re: a dozen cpu's on a chip |
On Thu, 08 May 2008 18:26:05 -0700, John Larkin <jjlarkin@highNOTlandTHIStechnologyPART.com > wrote: >On Thu, 08 May 2008 20:42:42 GMT, nico@puntnl.niks (Nico Coesel) >wrote: > >>John Larkin <jjlarkin@highNOTlandTHIStechnologyPART.com> wrote: >> >>> >>> >>>http://www.eetimes.com/news/latest/showArticle.jhtml;jsessionid=CESEXHWXUFJNKQSNDLSCKHA?articleID=207600531 >>> >>> >>>I bet we'll see 256 one of these days. >> >>I doubt it. A mailman won't deliver mail faster when he is using a >>Ferrari instead of bicycle. I think we are going to see slower, >>cheaper, more energy efficient computers with back-to-basic software >>for every day / office use. > >Maybe. But there's certainly no trend in that direction yet. Vista is >even more bloated than XP. My conjecture is that multiple cpu's on one >chip will make computers simpler and more efficient, and certainly >more reliable. > > >> >>For example an embedded ARM or MIPS system running at 300MHz has >>enough power to run most common applications. It consumes at least 10 >>times less energy compared to a standard PC. > >Oh, things are coming along nicely: > >http://www.sun.com/emrkt/innercircle/newsletter/0407feature.html > >"Sun foresees the need for extending thread count beyond 64 separate >instances of a computer, which is why the chip that will follow >UltraSPARC T2 (currently code-named Victoria Falls) is being designed >to have 128 threads. These threads, however, will be fully extendible. >This will make it possible to link two instances of Victoria Falls — >both sharing common memory through a hub chip — for a grand total of >256 threads." > >Just threads, not full cpu's [1], but it looks good. > >Here's one PPC and eight smaller processors on one chip: > >http://domino.research.ibm.com/comm/research_projects.nsf/pages/multicore.CellBE.html > > >John The Cell will be in our future. The power 6 looks pretty good too. A power 6 Cell would be neat. > >[1] I kept the apostrophe so that people whose expertise stops at 20 >KHz can have something clever to say. But to me, cpus looks awkward; >you'd pronounce it "seepuss." CPUs It's an acronym, so it should be capitalized. The rules on pluralizing an acronym are what need to be defined. |
| 08 May 2008 23:48:46 |
| Spehro Pefhany |
| Re: a dozen cpu's on a chip |
On Thu, 08 May 2008 23:01:24 +0100, the renowned Eeyore <rabbitsfriendsandrelations@hotmail.com > wrote: > > >Spehro Pefhany wrote: > >> Eeyore wrote: >> >Apostrophe Police wrote: >> >> Eeyore wrote: >> >> > John Larkin wrote: >> >> >> http://www.eetimes.com/news/latest/showArticle.jhtml;jsessionid=CESEXHWXUFJNKQSNDLSCKHA?articleID=207600531 >> >> >> >> >> >> I bet we'll see 256 one of these days. >> >> > >> >> > But when will we see you cease using the greengrocers' apostrophe ? >> >> > >> >> > http://en.wikipedia.org/wiki/Apostrophe#Greengrocers.27_apostrophes >> >> >> >> Where exactly in John's post did you see one of those? >> >> >> >> -- >> >> Rich Grise, Self-Appointed Chief, >> >> Apostrophe Police >> > >> >How about the subject line ? >> >> That's not one of them Greengrocers' apostrophe's > >Actually it is. See the article I linked to. A lovely example one sees even today in some British pubs on the menu is >"fish and chip's". > >Maybe it should now be called the 'publicans' apostrophe' ? There's few greengrocers still around these days. > >> -- rather that's a >> standard (now largely deprecated) way of pluralizing abbreviations. >> It's the method I was taught in school, and if you search the New York >> Times archives you'll find frequent examples up until a few years ago. >> http://en.wikipedia.org/wiki/Apostrophe > >A damn ex-colonial abberation of course ! ;~) > > >> "An apostrophe is used by some writers to form a plural for >> abbreviations, acronyms, and symbols where adding just s rather than >> 's may leave things ambiguous or inelegant. > >I've yet to see any written precedent for this. > > >> While British English >> formerly endorsed the use of such apostrophes after numbers and dates, >> this usage has now largely been superseded" > >Quite so. > > >> Personally, I think it's less ambiguous in some cases so I'd rather >> see it used for all cases, but I guess the (spit) prescriptivists have >> won this round. > >Eh ? > >Lets see; plurals should not be formed by the use of apostrophes. Seems simple enough to me. For example, for >clarity, rather than pcb's I use PCBs. > >Would you seriously think of talking of diode's, resistor's or transistor's ? > Certainly not, but NPN's and BJT's, and even 1N4148's all look right to me. And if I want to connect the Vss's together on two chips, is there any sensible alternative? Best regards, Spehro Pefhany -- "it's the network..." "The Journey is the reward" speff@interlog.com Info for manufacturers: http://www.trexon.com Embedded software/hardware/analog Info for designers: http://www.speff.com |
| 09 May 2008 09:09:38 |
| Martin Brown |
| Re: a dozen cpu's on a chip |
John Larkin wrote: > On Thu, 08 May 2008 20:42:42 GMT, nico@puntnl.niks (Nico Coesel) > wrote: > >> John Larkin <jjlarkin@highNOTlandTHIStechnologyPART.com> wrote: >> >>> http://www.eetimes.com/news/latest/showArticle.jhtml;jsessionid=CESEXHWXUFJNKQSNDLSCKHA?articleID=207600531 >>> >>> I bet we'll see 256 one of these days. I'll bet we won't. Except maybe as an educational design excercise. Multiple CPUs are hard to manage efficiently for general purpose computation when N >4. N<=3 is pretty trivial, 4 was interesting with the Chinese usage (ie big iron box didn't work and had service engineers tending it daily for about 6 months after delivery). >> I doubt it. A mailman won't deliver mail faster when he is using a >> Ferrari instead of bicycle. I think we are going to see slower, >> cheaper, more energy efficient computers with back-to-basic software >> for every day / office use. > > Maybe. But there's certainly no trend in that direction yet. Yes there is and they are all around you in consumer items like mobile phones and PDAs. ARM has just announced their multicore CPUs end of last year designed to have performance on demand for streaming video whilst having extremely low power consumption when doing ordinary GUI stuff. eg. http://www.itweek.co.uk/itweek/news/2200698/arm-unveils-multi-core-phone > Vista is even more bloated than XP. No denying that, but it is from Mickeysoft so what do you expect. XL2007 is barely functional but still business users flock to buy it like sheep. > My conjecture is that multiple cpu's on one > chip will make computers simpler and more efficient, and certainly > more reliable. It will do none of the above. Systems using multiple CPUs have been tried already using separate CPUs. I doubt if common consumer kit will ever go beyond 4 cores. Law of diminishing returns sets in about there. Multiple CPUs are good for certain types of problem requiring the same calculation on lots of data or a calculation that can be split across multiple threads but they are not well suited to general computing. >> For example an embedded ARM or MIPS system running at 300MHz has >> enough power to run most common applications. It consumes at least 10 >> times less energy compared to a standard PC. > > Oh, things are coming along nicely: > > http://www.sun.com/emrkt/innercircle/newsletter/0407feature.html > > "Sun foresees the need for extending thread count beyond 64 separate > instances of a computer, which is why the chip that will follow > UltraSPARC T2 (currently code-named Victoria Falls) is being designed > to have 128 threads. These threads, however, will be fully extendible. > This will make it possible to link two instances of Victoria Falls — > both sharing common memory through a hub chip — for a grand total of > 256 threads." > > Just threads, not full cpu's [1], but it looks good. Hardware support for threads and context switching makes good sense, but it isn't all that exciting unless you have a thing about lots of CPUs. Regards, Martin Brown ** Posted from http://www.teranews.com** |
| 09 May 2008 11:28:06 |
| TheM |
| Re: a dozen cpu's on a chip |
"Martin Brown" <|||newspam|||@nezumi.demon.co.uk > wrote in message news:4145e$482406c2$13322@news.teranews.com... >>>> I bet we'll see 256 one of these days. > > I'll bet we won't. Except maybe as an educational design excercise. > Multiple CPUs are hard to manage efficiently for general purpose computation when N>4. N<=3 is pretty trivial, 4 was interesting > with the Chinese usage (ie big iron box didn't work and had service engineers tending it daily for about 6 months after delivery). Think again. AMD is planning 12 cores for their next wunderchip.That's right, not 8 but 12. Probably due to the fact that their cores now don't perform on par with Intel's so this is how they plan to catch up/prevail. If they don't fold before they get it out with their massive debts and losses. We'll probably see 5,6,7,8,9,10,11 cores if they attempt to sell their broken chips as they are doing now with their quads sold as 3-core processors. More than 4 will definitely happen (and soon) regardless if it makes sense or not simply because numbers sell. Marketing people want it. It will be interesting to see how far this game goes. Mark |
| 09 May 2008 11:18:39 |
| Martin Brown |
| Re: a dozen cpu's on a chip |
TheM wrote: > "Martin Brown" <|||newspam|||@nezumi.demon.co.uk> wrote in message news:4145e$482406c2$13322@news.teranews.com... >>>>> I bet we'll see 256 one of these days. >> I'll bet we won't. Except maybe as an educational design excercise. >> Multiple CPUs are hard to manage efficiently for general purpose computation when N>4. N<=3 is pretty trivial, 4 was interesting >> with the Chinese usage (ie big iron box didn't work and had service engineers tending it daily for about 6 months after delivery). > > Think again. AMD is planning 12 cores for their next wunderchip. There is an important difference between planning and delivering. > That's > right, not 8 but 12. Probably due to the fact that their cores now don't perform > on par with Intel's so this is how they plan to catch up/prevail. If they don't fold > before they get it out with their massive debts and losses. N >4 is dangerous territory if you want to make it all work reliably. Still they may go out in a blaze of glory persuing this goal. > We'll probably see 5,6,7,8,9,10,11 cores if they attempt to sell their broken > chips as they are doing now with their quads sold as 3-core processors. N >=4 is relatively easy. N>4 gets a lot stiffer. > > More than 4 will definitely happen (and soon) regardless if it makes sense or not > simply because numbers sell. Marketing people want it. > It will be interesting to see how far this game goes. Probably a bit like the megapixel race in digicams there will be a demented marketting push for ever larger N to the point where N is larger and the pixels smaller than anything that makes sense with the ancillary hardware. Regards, Martin Brown ** Posted from http://www.teranews.com** |
| 09 May 2008 07:15:38 |
| John Larkin |
| Re: a dozen cpu's on a chip |
On Fri, 09 May 2008 11:18:39 +0100, Martin Brown <|||newspam|||@nezumi.demon.co.uk > wrote: >TheM wrote: >> "Martin Brown" <|||newspam|||@nezumi.demon.co.uk> wrote in message news:4145e$482406c2$13322@news.teranews.com... >>>>>> I bet we'll see 256 one of these days. > >>> I'll bet we won't. Except maybe as an educational design excercise. >>> Multiple CPUs are hard to manage efficiently for general purpose computation when N>4. N<=3 is pretty trivial, 4 was interesting >>> with the Chinese usage (ie big iron box didn't work and had service engineers tending it daily for about 6 months after delivery). >> >> Think again. AMD is planning 12 cores for their next wunderchip. > >There is an important difference between planning and delivering. > >> That's >> right, not 8 but 12. Probably due to the fact that their cores now don't perform >> on par with Intel's so this is how they plan to catch up/prevail. If they don't fold >> before they get it out with their massive debts and losses. > >N>4 is dangerous territory if you want to make it all work reliably. >Still they may go out in a blaze of glory persuing this goal. > >> We'll probably see 5,6,7,8,9,10,11 cores if they attempt to sell their broken >> chips as they are doing now with their quads sold as 3-core processors. > >N>=4 is relatively easy. N>4 gets a lot stiffer. You seem to be assuming that one program will be run on multiple cores for performance. Yes, that's difficult. But it's unnecessary, since the average user doesn't do FFTs all day; he edits, browses, prints. If one cpu is allocated to each process, everything becomes simple, and we avoid context switching and, with a decent system design, a crash-proof OS can be easily designed. It's not about performance; it's about escaping a programming methodology that's clearly broken. John |
| 09 May 2008 07:42:29 |
| MooseFET |
| Re: a dozen cpu's on a chip |
On May 8, 8:27 am, John Larkin <jjlar...@highNOTlandTHIStechnologyPART.com > wrote: > On Thu, 8 May 2008 07:42:04 -0700 (PDT), MooseFET <kensm...@rahul.net> > wrote: > > >On May 7, 7:48 pm, John Larkin > ><jjlar...@highNOTlandTHIStechnologyPART.com> wrote: > >>http://www.eetimes.com/news/latest/showArticle.jhtml;jsessionid=CESEX... > > >> I bet we'll see 256 one of these days. > > >When you get to large numbers of CPUs it seems to make sense to stop > >making them identical. For servers this would be doubly so. Many of > >the CPUs won't need to do floating point operations. > > Right. Amybe a few cpu's would have serious floating point power, or a > few separate fp engines could be assigned to cpu's as needed. Lots of > cpu's, doing stuff like file i/o or serial stuff, could be less > powerful. I suppose we'll always need special graphics hardware, but > just a few of those per chip. It could go even further. You could have a situation where the "boss" integer only CPU does this: Dear Mr Floating processor #1: Please go perform the code at the following address. Hey Byte slinger processor #7: Go make this memory move. Hey I/O processor #3: go do this work. ... etc .... Early in the era of the 8086 there was an 8089 which was called a DMA processor even though it really was programmed I/O in its own instruction set. It could do I/O way faster than the 8086. If a CPU is intended to be part of a server, it could have parts like that in it for doing the things needed for fast disk operations. > >It also would make sense to do things like memory moves in the "Memory > >Mismanagement Unit" since the values don't need to be modified on the > >way through. > > >This will make it a lot harder to say how many CPUs are in a chip. If > >there is only as much hardware as 200 full CPUs but 500 threads can be > >running at the same time, do you call it 200 or 500 CPUs. > > Next step is to get rid of task swapping and threads altogether. One > CPU is the OS, and one cpu gets assigned per process. Actually I can see multiple CPUs being the OS. When you have a lot of tasks to manage, it takes some processing power just to manage them. Future machines with thousands of CPUs may need tens of managers. > > John |
| 09 May 2008 15:48:22 |
| Martin Brown |
| Re: a dozen cpu's on a chip |
John Larkin wrote: > On Fri, 09 May 2008 11:18:39 +0100, Martin Brown > <|||newspam|||@nezumi.demon.co.uk> wrote: > >> TheM wrote: >>> "Martin Brown" <|||newspam|||@nezumi.demon.co.uk> wrote in message news:4145e$482406c2$13322@news.teranews.com... >>>>>>> I bet we'll see 256 one of these days. >>>> I'll bet we won't. Except maybe as an educational design excercise. >>>> Multiple CPUs are hard to manage efficiently for general purpose computation when N>4. N<=3 is pretty trivial, 4 was interesting >>>> with the Chinese usage (ie big iron box didn't work and had service engineers tending it daily for about 6 months after delivery). >>> Think again. AMD is planning 12 cores for their next wunderchip. >> There is an important difference between planning and delivering. >> N>4 is dangerous territory if you want to make it all work reliably. >> Still they may go out in a blaze of glory persuing this goal. >> >>> We'll probably see 5,6,7,8,9,10,11 cores if they attempt to sell their broken >>> chips as they are doing now with their quads sold as 3-core processors. >> N>=4 is relatively easy. N>4 gets a lot stiffer. NB Typo should be N<=4 is relatively easy. > You seem to be assuming that one program will be run on multiple cores > for performance. No. I was actually thinking of hardware problems related to shared common memory, bus arbitration, bandwidth and cache coherency. > Yes, that's difficult. But it's unnecessary, since > the average user doesn't do FFTs all day; he edits, browses, prints. All of which are easily done by timeslicing a single CPU. > If one cpu is allocated to each process, everything becomes simple, > and we avoid context switching and, with a decent system design, a > crash-proof OS can be easily designed. Dream on. It isn't the context switching that is the problem it is bad or no design, sloppy programming languages and sloppy programmers. OS/2 despite being designed by committee and reinventing a few wheels was going in the right direction and later versions were very close to bulletproof. The early marketting decision to hobble it to be backwards compatible with the 286 CPU messed things up. Unfortunately IBM & Mickeysoft fell out and pure 386 Windows gained ascendency. > > It's not about performance; it's about escaping a programming > methodology that's clearly broken. Having spare CPUs for some tasks is quite handy, but there is a law of diminishing returns for N >4 unless your problem has special properties. My PC is analysing a chess game with another CPU as I type this. Apart from the additional fan noise there is no performance hit. If I allowed it to use both CPUs the keyboard lag would be painful. Regards, Martin Brown ** Posted from http://www.teranews.com** |
| 09 May 2008 07:52:57 |
| gearhead |
| Re: a dozen cpu's on a chip |
On May 8, 6:41=A0pm, FatBytestard <FatBytest...@somewheronyourharddrive.org > wrote: > On Thu, 08 May 2008 18:26:05 -0700, John Larkin > > > > > > <jjlar...@highNOTlandTHIStechnologyPART.com> wrote: > >On Thu, 08 May 2008 20:42:42 GMT, n...@puntnl.niks (Nico Coesel) > >wrote: > > >>John Larkin <jjlar...@highNOTlandTHIStechnologyPART.com> wrote: > > >>>http://www.eetimes.com/news/latest/showArticle.jhtml;jsessionid=3DCESEX= ... > > >>>I bet we'll see 256 one of these days. > > >>I doubt it. A mailman won't deliver mail faster when he is using a > >>Ferrari instead of bicycle. I think we are going to see slower, > >>cheaper, more energy efficient computers with back-to-basic software > >>for every day / office use. > > >Maybe. But there's certainly no trend in that direction yet. Vista is > >even more bloated than XP. My conjecture is that multiple cpu's on one > >chip will make computers simpler and more efficient, and certainly > >more reliable. > > >>For example an embedded ARM or MIPS system running at 300MHz has > >>enough power to run most common applications. It consumes at least 10 > >>times less energy compared to a standard PC. > > >Oh, things are coming along nicely: > > >http://www.sun.com/emrkt/innercircle/newsletter/0407feature.html > > >"Sun foresees the need for extending thread count beyond 64 separate > >instances of a computer, which is why the chip that will follow > >UltraSPARC T2 (currently code-named Victoria Falls) is being designed > >to have 128 threads. These threads, however, will be fully extendible. > >This will make it possible to link two instances of Victoria Falls =97 > >both sharing common memory through a hub chip =97 for a grand total of > >256 threads." > > >Just threads, not full cpu's [1], but it looks good. > > >Here's one PPC and eight smaller processors on one chip: > > >http://domino.research.ibm.com/comm/research_projects.nsf/pages/multi... > > >John > > =A0 The Cell will be in our future. =A0The power 6 looks pretty good too. > > =A0 A power 6 Cell would be neat. > > > > >[1] I kept the apostrophe so that people whose expertise stops at 20 > >KHz can have something clever to say. But to me, cpus looks awkward; > >you'd pronounce it "seepuss." > > =A0CPUs =A0It's an acronym, so it should be capitalized. The rules on > pluralizing an acronym are what need to be defined.- Hide quoted text - > > - Show quoted text - The rule: for acronyms of five letters or more, capitalize only the first letter (e.g., Unicef). |
| 09 May 2008 08:04:03 |
| Jim Thompson |
| Re: a dozen cpu's on a chip |
On Fri, 9 May 2008 07:52:57 -0700 (PDT), gearhead <nospam@billburg.com > wrote: >On May 8, 6:41 pm, FatBytestard ><FatBytest...@somewheronyourharddrive.org> wrote: >> On Thu, 08 May 2008 18:26:05 -0700, John Larkin >> >> >> >> >> >> <jjlar...@highNOTlandTHIStechnologyPART.com> wrote: >> >On Thu, 08 May 2008 20:42:42 GMT, n...@puntnl.niks (Nico Coesel) >> >wrote: >> >> >>John Larkin <jjlar...@highNOTlandTHIStechnologyPART.com> wrote: >> >> >>>http://www.eetimes.com/news/latest/showArticle.jhtml;jsessionid=CESEX... >> >> >>>I bet we'll see 256 one of these days. >> >> >>I doubt it. A mailman won't deliver mail faster when he is using a >> >>Ferrari instead of bicycle. I think we are going to see slower, >> >>cheaper, more energy efficient computers with back-to-basic software >> >>for every day / office use. >> >> >Maybe. But there's certainly no trend in that direction yet. Vista is >> >even more bloated than XP. My conjecture is that multiple cpu's on one >> >chip will make computers simpler and more efficient, and certainly >> >more reliable. >> >> >>For example an embedded ARM or MIPS system running at 300MHz has >> >>enough power to run most common applications. It consumes at least 10 >> >>times less energy compared to a standard PC. >> >> >Oh, things are coming along nicely: >> >> >http://www.sun.com/emrkt/innercircle/newsletter/0407feature.html >> >> >"Sun foresees the need for extending thread count beyond 64 separate >> >instances of a computer, which is why the chip that will follow >> >UltraSPARC T2 (currently code-named Victoria Falls) is being designed >> >to have 128 threads. These threads, however, will be fully extendible. >> >This will make it possible to link two instances of Victoria Falls — >> >both sharing common memory through a hub chip — for a grand total of >> >256 threads." >> >> >Just threads, not full cpu's [1], but it looks good. >> >> >Here's one PPC and eight smaller processors on one chip: >> >> >http://domino.research.ibm.com/comm/research_projects.nsf/pages/multi... >> >> >John >> >> The Cell will be in our future. The power 6 looks pretty good too. >> >> A power 6 Cell would be neat. >> >> >> >> >[1] I kept the apostrophe so that people whose expertise stops at 20 >> >KHz can have something clever to say. But to me, cpus looks awkward; >> >you'd pronounce it "seepuss." >> >> CPUs It's an acronym, so it should be capitalized. The rules on >> pluralizing an acronym are what need to be defined.- Hide quoted text - >> >> - Show quoted text - > >The rule: for acronyms of five letters or more, capitalize only the >first letter >(e.g., Unicef). Is there really a "rule" ?:-) ...Jim Thompson -- | James E.Thompson, P.E. | mens | | Analog Innovations, Inc. | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | Phoenix, Arizona 85048 Skype: "skypeanalog" | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at http://www.analog-innovations.com| 1962 | America: Land of the Free, Because of the Brave |
| 09 May 2008 08:42:30 |
| gearhead |
| Re: a dozen cpu's on a chip |
On May 9, 8:04=A0am, Jim Thompson <To-Email-Use-The-Envelope-I...@My-Web- Site.com > wrote: > On Fri, 9 May 2008 07:52:57 -0700 (PDT), gearhead > > > > > > <nos...@billburg.com> wrote: > >On May 8, 6:41=A0pm, FatBytestard > ><FatBytest...@somewheronyourharddrive.org> wrote: > >> On Thu, 08 May 2008 18:26:05 -0700, John Larkin > > >> <jjlar...@highNOTlandTHIStechnologyPART.com> wrote: > >> >On Thu, 08 May 2008 20:42:42 GMT, n...@puntnl.niks (Nico Coesel) > >> >wrote: > > >> >>John Larkin <jjlar...@highNOTlandTHIStechnologyPART.com> wrote: > > >> >>>http://www.eetimes.com/news/latest/showArticle.jhtml;jsessionid=3DCE= SEX... > > >> >>>I bet we'll see 256 one of these days. > > >> >>I doubt it. A mailman won't deliver mail faster when he is using a > >> >>Ferrari instead of bicycle. I think we are going to see slower, > >> >>cheaper, more energy efficient computers with back-to-basic software > >> >>for every day / office use. > > >> >Maybe. But there's certainly no trend in that direction yet. Vista is > >> >even more bloated than XP. My conjecture is that multiple cpu's on one= > >> >chip will make computers simpler and more efficient, and certainly > >> >more reliable. > > >> >>For example an embedded ARM or MIPS system running at 300MHz has > >> >>enough power to run most common applications. It consumes at least 10= > >> >>times less energy compared to a standard PC. > > >> >Oh, things are coming along nicely: > > >> >http://www.sun.com/emrkt/innercircle/newsletter/0407feature.html > > >> >"Sun foresees the need for extending thread count beyond 64 separate > >> >instances of a computer, which is why the chip that will follow > >> >UltraSPARC T2 (currently code-named Victoria Falls) is being designed > >> >to have 128 threads. These threads, however, will be fully extendible.= > >> >This will make it possible to link two instances of Victoria Falls =97= > >> >both sharing common memory through a hub chip =97 for a grand total of= > >> >256 threads." > > >> >Just threads, not full cpu's [1], but it looks good. > > >> >Here's one PPC and eight smaller processors on one chip: > > >> >http://domino.research.ibm.com/comm/research_projects.nsf/pages/multi.= .. > > >> >John > > >> =A0 The Cell will be in our future. =A0The power 6 looks pretty good to= o. > > >> =A0 A power 6 Cell would be neat. > > >> >[1] I kept the apostrophe so that people whose expertise stops at 20 > >> >KHz can have something clever to say. But to me, cpus looks awkward; > >> >you'd pronounce it "seepuss." > > >> =A0CPUs =A0It's an acronym, so it should be capitalized. The rules on > >> pluralizing an acronym are what need to be defined.- Hide quoted text -= > > >> - Show quoted text - > > >The rule: =A0for acronyms of five letters or more, capitalize only the > >first letter > >(e.g., Unicef). > > Is there really a "rule" ?:-) > > =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 = =A0 =A0 ...Jim Thompson Yes, and if you break it Eeyore will spank you. |
| 09 May 2008 08:49:29 |
| John Larkin |
| Re: a dozen cpu's on a chip |
On Fri, 9 May 2008 07:42:29 -0700 (PDT), MooseFET <kensmith@rahul.net > wrote: >On May 8, 8:27 am, John Larkin ><jjlar...@highNOTlandTHIStechnologyPART.com> wrote: >> On Thu, 8 May 2008 07:42:04 -0700 (PDT), MooseFET <kensm...@rahul.net> >> wrote: >> >> >On May 7, 7:48 pm, John Larkin >> ><jjlar...@highNOTlandTHIStechnologyPART.com> wrote: >> >>http://www.eetimes.com/news/latest/showArticle.jhtml;jsessionid=CESEX... >> >> >> I bet we'll see 256 one of these days. >> >> >When you get to large numbers of CPUs it seems to make sense to stop >> >making them identical. For servers this would be doubly so. Many of >> >the CPUs won't need to do floating point operations. >> >> Right. Amybe a few cpu's would have serious floating point power, or a >> few separate fp engines could be assigned to cpu's as needed. Lots of >> cpu's, doing stuff like file i/o or serial stuff, could be less >> powerful. I suppose we'll always need special graphics hardware, but >> just a few of those per chip. > >It could go even further. You could have a situation where the "boss" >integer only CPU does this: > >Dear Mr Floating processor #1: Please go perform the code at the >following address. > >Hey Byte slinger processor #7: Go make this memory move. > >Hey I/O processor #3: go do this work. > >... etc .... Yes. But I think that stuff like FPU sharing could be done in hardware, with the OS cpu just setting up the rules. Really scary stuff line memory sharing/moves, disk writes, and DMA - things that work across all physical memory - could only be done by a manager cpu, never by an application. VMS had some protective features like that, and ran dozens of hostile users for months at a whack. > >Early in the era of the 8086 there was an 8089 which was called a DMA >processor even though it really was programmed I/O in its own >instruction set. It could do I/O way faster than the 8086. If a CPU >is intended to be part of a server, it could have parts like that in >it for doing the things needed for fast disk operations. > > > > >> >It also would make sense to do things like memory moves in the "Memory >> >Mismanagement Unit" since the values don't need to be modified on the >> >way through. >> >> >This will make it a lot harder to say how many CPUs are in a chip. If >> >there is only as much hardware as 200 full CPUs but 500 threads can be >> >running at the same time, do you call it 200 or 500 CPUs. >> >> Next step is to get rid of task swapping and threads altogether. One >> CPU is the OS, and one cpu gets assigned per process. > >Actually I can see multiple CPUs being the OS. When you have a lot of >tasks to manage, it takes some processing power just to manage them. >Future machines with thousands of CPUs may need tens of managers. But only one executive manager boss. It could be a structured heirarchy: one cpu runs the top-level OS, the thing that gets booted and sets up/supervises everybody else; a few file managers; a bunch of hardware device drivers; a user/GUI interface cpu per screen; some TCP/IP managers, one per port maybe; finally at the bottom of the stack would be applications, which would be rigidly hardware-constrained from doing harm. The only thing wrong with running all this one a single shared cpu is that nobody seems to be able to do it right. The multi-cpu thing isn't about performance or even architecture: it's a way to force programmers into a new set of rules, because they sure can't manage the ones they have now. Oh, yeah, dump virtual memory. It was a bad idea when memory was scarce, and a worse one now that ram is bucks per gigabyte. John |
| 09 May 2008 08:50:42 |
| John Larkin |
| Re: a dozen cpu's on a chip |
On Fri, 09 May 2008 08:04:03 -0700, Jim Thompson <To-Email-Use-The-Envelope-Icon@My-Web-Site.com > wrote: >On Fri, 9 May 2008 07:52:57 -0700 (PDT), gearhead ><nospam@billburg.com> wrote: > >>On May 8, 6:41 pm, FatBytestard >><FatBytest...@somewheronyourharddrive.org> wrote: >>> On Thu, 08 May 2008 18:26:05 -0700, John Larkin >>> >>> >>> >>> >>> >>> <jjlar...@highNOTlandTHIStechnologyPART.com> wrote: >>> >On Thu, 08 May 2008 20:42:42 GMT, n...@puntnl.niks (Nico Coesel) >>> >wrote: >>> >>> >>John Larkin <jjlar...@highNOTlandTHIStechnologyPART.com> wrote: >>> >>> >>>http://www.eetimes.com/news/latest/showArticle.jhtml;jsessionid=CESEX... >>> >>> >>>I bet we'll see 256 one of these days. >>> >>> >>I doubt it. A mailman won't deliver mail faster when he is using a >>> >>Ferrari instead of bicycle. I think we are going to see slower, >>> >>cheaper, more energy efficient computers with back-to-basic software >>> >>for every day / office use. >>> >>> >Maybe. But there's certainly no trend in that direction yet. Vista is >>> >even more bloated than XP. My conjecture is that multiple cpu's on one >>> >chip will make computers simpler and more efficient, and certainly >>> >more reliable. >>> >>> >>For example an embedded ARM or MIPS system running at 300MHz has >>> >>enough power to run most common applications. It consumes at least 10 >>> >>times less energy compared to a standard PC. >>> >>> >Oh, things are coming along nicely: >>> >>> >http://www.sun.com/emrkt/innercircle/newsletter/0407feature.html >>> >>> >"Sun foresees the need for extending thread count beyond 64 separate >>> >instances of a computer, which is why the chip that will follow >>> >UltraSPARC T2 (currently code-named Victoria Falls) is being designed >>> >to have 128 threads. These threads, however, will be fully extendible. >>> >This will make it possible to link two instances of Victoria Falls — >>> >both sharing common memory through a hub chip — for a grand total of >>> >256 threads." >>> >>> >Just threads, not full cpu's [1], but it looks good. >>> >>> >Here's one PPC and eight smaller processors on one chip: >>> >>> >http://domino.research.ibm.com/comm/research_projects.nsf/pages/multi... >>> >>> >John >>> >>> The Cell will be in our future. The power 6 looks pretty good too. >>> >>> A power 6 Cell would be neat. >>> >>> >>> >>> >[1] I kept the apostrophe so that people whose expertise stops at 20 >>> >KHz can have something clever to say. But to me, cpus looks awkward; >>> >you'd pronounce it "seepuss." >>> >>> CPUs It's an acronym, so it should be capitalized. The rules on >>> pluralizing an acronym are what need to be defined.- Hide quoted text - >>> >>> - Show quoted text - >> >>The rule: for acronyms of five letters or more, capitalize only the >>first letter >>(e.g., Unicef). > >Is there really a "rule" ?:-) > > ...Jim Thompson I have only one rule: always make your own rules. John |
| 09 May 2008 08:52:42 |
| Jim Thompson |
| Re: a dozen cpu's on a chip |
On Fri, 9 May 2008 08:42:30 -0700 (PDT), gearhead <nospam@billburg.com > wrote: >On May 9, 8:04 am, Jim Thompson <To-Email-Use-The-Envelope-I...@My-Web- >Site.com> wrote: >> On Fri, 9 May 2008 07:52:57 -0700 (PDT), gearhead >> >> >> >> >> >> <nos...@billburg.com> wrote: >> >On May 8, 6:41 pm, FatBytestard >> ><FatBytest...@somewheronyourharddrive.org> wrote: >> >> On Thu, 08 May 2008 18:26:05 -0700, John Larkin >> >> >> <jjlar...@highNOTlandTHIStechnologyPART.com> wrote: >> >> >On Thu, 08 May 2008 20:42:42 GMT, n...@puntnl.niks (Nico Coesel) >> >> >wrote: >> >> >> >>John Larkin <jjlar...@highNOTlandTHIStechnologyPART.com> wrote: >> >> >> >>>http://www.eetimes.com/news/latest/showArticle.jhtml;jsessionid=CESEX... >> >> >> >>>I bet we'll see 256 one of these days. >> >> >> >>I doubt it. A mailman won't deliver mail faster when he is using a >> >> >>Ferrari instead of bicycle. I think we are going to see slower, >> >> >>cheaper, more energy efficient computers with back-to-basic software >> >> >>for every day / office use. >> >> >> >Maybe. But there's certainly no trend in that direction yet. Vista is >> >> >even more bloated than XP. My conjecture is that multiple cpu's on one >> >> >chip will make computers simpler and more efficient, and certainly >> >> >more reliable. >> >> >> >>For example an embedded ARM or MIPS system running at 300MHz has >> >> >>enough power to run most common applications. It consumes at least 10 >> >> >>times less energy compared to a standard PC. >> >> >> >Oh, things are coming along nicely: >> >> >> >http://www.sun.com/emrkt/innercircle/newsletter/0407feature.html >> >> >> >"Sun foresees the need for extending thread count beyond 64 separate >> >> >instances of a computer, which is why the chip that will follow >> >> >UltraSPARC T2 (currently code-named Victoria Falls) is being designed >> >> >to have 128 threads. These threads, however, will be fully extendible. >> >> >This will make it possible to link two instances of Victoria Falls — >> >> >both sharing common memory through a hub chip — for a grand total of >> >> >256 threads." >> >> >> >Just threads, not full cpu's [1], but it looks good. >> >> >> >Here's one PPC and eight smaller processors on one chip: >> >> >> >http://domino.research.ibm.com/comm/research_projects.nsf/pages/multi... >> >> >> >John >> >> >> The Cell will be in our future. The power 6 looks pretty good too. >> >> >> A power 6 Cell would be neat. >> >> >> >[1] I kept the apostrophe so that people whose expertise stops at 20 >> >> >KHz can have something clever to say. But to me, cpus looks awkward; >> >> >you'd pronounce it "seepuss." >> >> >> CPUs It's an acronym, so it should be capitalized. The rules on >> >> pluralizing an acronym are what need to be defined.- Hide quoted text - >> >> >> - Show quoted text - >> >> >The rule: for acronyms of five letters or more, capitalize only the >> >first letter >> >(e.g., Unicef). >> >> Is there really a "rule" ?:-) >> >> ...Jim Thompson > >Yes, and if you break it Eeyore will spank you. Who is this Eeyore? I don't see any posts from him, or any responders (troll feeders) to his posts, so I guess I don't need to concern myself :-) ...Jim Thompson -- | James E.Thompson, P.E. | mens | | Analog Innovations, Inc. | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | Phoenix, Arizona 85048 Skype: "skypeanalog" | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at http://www.analog-innovations.com| 1962 | America: Land of the Free, Because of the Brave |
| 09 May 2008 09:08:12 |
| John Larkin |
| Re: a dozen cpu's on a chip |
On Fri, 09 May 2008 15:48:22 +0100, Martin Brown <|||newspam|||@nezumi.demon.co.uk > wrote: >John Larkin wrote: >> On Fri, 09 May 2008 11:18:39 +0100, Martin Brown >> <|||newspam|||@nezumi.demon.co.uk> wrote: >> >>> TheM wrote: >>>> "Martin Brown" <|||newspam|||@nezumi.demon.co.uk> wrote in message news:4145e$482406c2$13322@news.teranews.com... >>>>>>>> I bet we'll see 256 one of these days. > >>>>> I'll bet we won't. Except maybe as an educational design excercise. >>>>> Multiple CPUs are hard to manage efficiently for general purpose computation when N>4. N<=3 is pretty trivial, 4 was interesting >>>>> with the Chinese usage (ie big iron box didn't work and had service engineers tending it daily for about 6 months after delivery). > >>>> Think again. AMD is planning 12 cores for their next wunderchip. >>> There is an important difference between planning and delivering. > >>> N>4 is dangerous territory if you want to make it all work reliably. >>> Still they may go out in a blaze of glory persuing this goal. >>> >>>> We'll probably see 5,6,7,8,9,10,11 cores if they attempt to sell their broken >>>> chips as they are doing now with their quads sold as 3-core processors. >>> N>=4 is relatively easy. N>4 gets a lot stiffer. > >NB Typo should be N<=4 is relatively easy. > >> You seem to be assuming that one program will be run on multiple cores >> for performance. > >No. I was actually thinking of hardware problems related to shared >common memory, bus arbitration, bandwidth and cache coherency. > >> Yes, that's difficult. But it's unnecessary, since >> the average user doesn't do FFTs all day; he edits, browses, prints. > >All of which are easily done by timeslicing a single CPU. If it's easy, why are "home pc" OS's flakey gigabyte monsters that even professionals don't understand and struggle to maintain? And why would *any* OS be subject to acquiring a virus by just viewing a web page? And why does doing routine operations sometimes hang all tasks for tens of seconds? I'm just not seeing the "easily done" part; the Mac OS's come closest, I guess. > >> If one cpu is allocated to each process, everything becomes simple, >> and we avoid context switching and, with a decent system design, a >> crash-proof OS can be easily designed. > >Dream on. It isn't the context switching that is the problem it is bad >or no design, sloppy programming languages and sloppy programmers. Yes. But the only thing that will break the paradigm is a totally new hardware architecture. The multiple cpu's could be virtual (ie, a register that selects the "current cpu", and all of its contexts, instantly) but multiple cores with true concurrent execution is probably a better use of silicon and a better way to manage resources, especially the most difficult resource: time [1]. The point is that new hardware is needed to constrain bad programmers (ie, most programmers) from trying to do complex, clever things that they rarely do right. A cpu is a tiny spec of silicon these days. John [1] Could it be that programmers tend to look at programs as linearly executed screens of text, and realtime things - task switching, interrupts, async events - don't fit into their way of thinking? Maybe that's why the best programmers are people who were never educated as programmers. |
| 09 May 2008 09:17:23 |
| Joel Koltner |
| Re: a dozen cpu's on a chip |
<MassiveProng@thebarattheendoftheuniverse.org > wrote in message news:8ba724l5m6ql8mmg2bkq7ob3ioo2b4d9br@4ax.com... > Funny, my Vista installation is still working flawlessly, and any error > it ever did have was repaired either in session or on the next boot. I have a Vista-based laptop that has blue-screened twice now, and on both occasions it corrupted something in the initial boot bits of Windows such that it refused to boot anymore. Running the automated repair routine from the install CD did fix it, but what a pain in the arse! > I had one blue screen on boot, and it worked immediately after, and > found the issue to be an Nvidia driver, which nvidia replaced with a > newer, proper driver. A lot of machines from, e.g., Dell, Asus, etc. use "tweaked" drivers and often can't be replaced with newer ones from nvidia.com... yet the ones from the OEM lack the nVidia ones by many months. (I realize you can just head on over to laptopvideo2go.com and download a modified .inf, but that's something the average user isn't aware of.) ---Joel |
| 09 May 2008 09:18:34 |
| Joel Koltner |
| Re: a dozen cpu's on a chip |
"John Larkin" <jjlarkin@highNOTlandTHIStechnologyPART.com > wrote in message news:6ls824togt4bl2a0slun6nmupd6agso7md@4ax.com... > I have only one rule: always make your own rules. Mine is closer to, "special rules for special people." |
| 09 May 2008 10:20:40 |
| Don Bowey |
| Re: a dozen cpu's on a chip |
On 5/9/08 9:08 AM, in article kos824lnlilmk5pn9v5t698ga9li0f02bc@4ax.com, "John Larkin" <jjlarkin@highNOTlandTHIStechnologyPART.com > wrote: > On Fri, 09 May 2008 15:48:22 +0100, Martin Brown > <|||newspam|||@nezumi.demon.co.uk> wrote: > >> John Larkin wrote: >>> On Fri, 09 May 2008 11:18:39 +0100, Martin Brown >>> <|||newspam|||@nezumi.demon.co.uk> wrote: >>> >>>> TheM wrote: >>>>> "Martin Brown" <|||newspam|||@nezumi.demon.co.uk> wrote in message >>>>> news:4145e$482406c2$13322@news.teranews.com... >>>>>>>>> I bet we'll see 256 one of these days. >> >>>>>> I'll bet we won't. Except maybe as an educational design excercise. >>>>>> Multiple CPUs are hard to manage efficiently for general purpose >>>>>> computation when N>4. N<=3 is pretty trivial, 4 was interesting >>>>>> with the Chinese usage (ie big iron box didn't work and had service >>>>>> engineers tending it daily for about 6 months after delivery). >> >>>>> Think again. AMD is planning 12 cores for their next wunderchip. >>>> There is an important difference between planning and delivering. >> >>>> N>4 is dangerous territory if you want to make it all work reliably. >>>> Still they may go out in a blaze of glory persuing this goal. >>>> >>>>> We'll probably see 5,6,7,8,9,10,11 cores if they attempt to sell their >>>>> broken >>>>> chips as they are doing now with their quads sold as 3-core processors. >>>> N>=4 is relatively easy. N>4 gets a lot stiffer. >> >> NB Typo should be N<=4 is relatively easy. >> >>> You seem to be assuming that one program will be run on multiple cores >>> for performance. >> >> No. I was actually thinking of hardware problems related to shared >> common memory, bus arbitration, bandwidth and cache coherency. >> >>> Yes, that's difficult. But it's unnecessary, since >>> the average user doesn't do FFTs all day; he edits, browses, prints. >> >> All of which are easily done by timeslicing a single CPU. > > If it's easy, why are "home pc" OS's flakey gigabyte monsters that > even professionals don't understand and struggle to maintain? And why > would *any* OS be subject to acquiring a virus by just viewing a web > page? And why does doing routine operations sometimes hang all tasks > for tens of seconds? > > I'm just not seeing the "easily done" part; the Mac OS's come closest, > I guess. Mac OS X and MS XP seem to me to be equally stable. My wife (XP on Fujitsu laptop) is an avid surfer and downloader of various file types, but, except for the new applications I install for her, the only executable files are MS updates. It is about 6 years old and I have never had to anything more than clean-up the HD. I currently have one computer (Compaq) running XP, and one with OS X and XP. Both of them run as clean so far. These computers are exposed to a variety of executable files as my whimsy dictates. I wish I understood more about how some people have continuous dire problems that we manage to avoid. > >> >>> If one cpu is allocated to each process, everything becomes simple, >>> and we avoid context switching and, with a decent system design, a >>> crash-proof OS can be easily designed. >> >> Dream on. It isn't the context switching that is the problem it is bad >> or no design, sloppy programming languages and sloppy programmers. > > Yes. But the only thing that will break the paradigm is a totally new > hardware architecture. The multiple cpu's could be virtual (ie, a > register that selects the "current cpu", and all of its contexts, > instantly) but multiple cores with true concurrent execution is > probably a better use of silicon and a better way to manage resources, > especially the most difficult resource: time [1]. The point is that > new hardware is needed to constrain bad programmers (ie, most > programmers) from trying to do complex, clever things that they rarely > do right. A cpu is a tiny spec of silicon these days. > > > John > > [1] Could it be that programmers tend to look at programs as linearly > executed screens of text, and realtime things - task switching, > interrupts, async events - don't fit into their way of thinking? Maybe > that's why the best programmers are people who were never educated as > programmers. > > |
| 09 May 2008 18:51:23 |
| john jardine |
| Re: a dozen cpu's on a chip |
"John Larkin" <jjlarkin@highNOTlandTHIStechnologyPART.com > wrote in message news:kos824lnlilmk5pn9v5t698ga9li0f02bc@4ax.com... [...] > [1] Could it be that programmers tend to look at programs as linearly > executed screens of text, and realtime things - task switching, > interrupts, async events - don't fit into their way of thinking? Maybe > that's why the best programmers are people who were never educated as > programmers. A programmer friend was yesterday showing me a shiny new book he'd just bought to assist in some PC programming. It's a big Yellow thing with 1400 pages dealing with C# and .NET. He then took great amusement in randomly quoting inanities contained within. Point is, as a 'self trained' hardware, C and assembler programmer, he understood and could see the book for what it really is. Very few 'computer science' graduates are capable of making this distinction. |
| 09 May 2008 11:06:21 |
| Jim Thompson |
| Re: a dozen cpu's on a chip |
On Fri, 9 May 2008 18:51:23 +0100, "john jardine" <john.jardine@idnet.co.uk > wrote: > >"John Larkin" <jjlarkin@highNOTlandTHIStechnologyPART.com> wrote in message >news:kos824lnlilmk5pn9v5t698ga9li0f02bc@4ax.com... >[...] > >> [1] Could it be that programmers tend to look at programs as linearly >> executed screens of text, and realtime things - task switching, >> interrupts, async events - don't fit into their way of thinking? Maybe >> that's why the best programmers are people who were never educated as >> programmers. > > A programmer friend was yesterday showing me a shiny new book he'd just >bought to assist in some PC programming. >It's a big Yellow thing with 1400 pages dealing with C# and .NET. He then >took great amusement in randomly quoting inanities contained within. >Point is, as a 'self trained' hardware, C and assembler programmer, he >understood and could see the book for what it really is. >Very few 'computer science' graduates are capable of making this >distinction. > > From experience with two of my offspring, good programming seems to come from good language skills (daughter and son are both multi-lingual), irrespective of training in "computer science" (a term I consider an oxymoron, right up there with "political science"). ...Jim Thompson -- | James E.Thompson, P.E. | mens | | Analog Innovations, Inc. | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | Phoenix, Arizona 85048 Skype: "skypeanalog" | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at http://www.analog-innovations.com| 1962 | America: Land of the Free, Because of the Brave |
| 09 May 2008 11:52:59 |
| John Larkin |
| Re: a dozen cpu's on a chip |
On Fri, 09 May 2008 10:20:40 -0700, Don Bowey <dbowey@comcast.net > wrote: >On 5/9/08 9:08 AM, in article kos824lnlilmk5pn9v5t698ga9li0f02bc@4ax.com, >"John Larkin" <jjlarkin@highNOTlandTHIStechnologyPART.com> wrote: > >> On Fri, 09 May 2008 15:48:22 +0100, Martin Brown >> <|||newspam|||@nezumi.demon.co.uk> wrote: >> >>> John Larkin wrote: >>>> On Fri, 09 May 2008 11:18:39 +0100, Martin Brown >>>> <|||newspam|||@nezumi.demon.co.uk> wrote: >>>> >>>>> TheM wrote: >>>>>> "Martin Brown" <|||newspam|||@nezumi.demon.co.uk> wrote in message >>>>>> news:4145e$482406c2$13322@news.teranews.com... >>>>>>>>>> I bet we'll see 256 one of these days. >>> >>>>>>> I'll bet we won't. Except maybe as an educational design excercise. >>>>>>> Multiple CPUs are hard to manage efficiently for general purpose >>>>>>> computation when N>4. N<=3 is pretty trivial, 4 was interesting >>>>>>> with the Chinese usage (ie big iron box didn't work and had service >>>>>>> engineers tending it daily for about 6 months after delivery). >>> >>>>>> Think again. AMD is planning 12 cores for their next wunderchip. >>>>> There is an important difference between planning and delivering. >>> >>>>> N>4 is dangerous territory if you want to make it all work reliably. >>>>> Still they may go out in a blaze of glory persuing this goal. >>>>> >>>>>> We'll probably see 5,6,7,8,9,10,11 cores if they attempt to sell their >>>>>> broken >>>>>> chips as they are doing now with their quads sold as 3-core processors. >>>>> N>=4 is relatively easy. N>4 gets a lot stiffer. >>> >>> NB Typo should be N<=4 is relatively easy. >>> >>>> You seem to be assuming that one program will be run on multiple cores >>>> for performance. >>> >>> No. I was actually thinking of hardware problems related to shared >>> common memory, bus arbitration, bandwidth and cache coherency. >>> >>>> Yes, that's difficult. But it's unnecessary, since >>>> the average user doesn't do FFTs all day; he edits, browses, prints. >>> >>> All of which are easily done by timeslicing a single CPU. >> >> If it's easy, why are "home pc" OS's flakey gigabyte monsters that >> even professionals don't understand and struggle to maintain? And why >> would *any* OS be subject to acquiring a virus by just viewing a web >> page? And why does doing routine operations sometimes hang all tasks >> for tens of seconds? >> >> I'm just not seeing the "easily done" part; the Mac OS's come closest, >> I guess. > >Mac OS X and MS XP seem to me to be equally stable. > >My wife (XP on Fujitsu laptop) is an avid surfer and downloader of various >file types, but, except for the new applications I install for her, the only >executable files are MS updates. It is about 6 years old and I have never >had to anything more than clean-up the HD. > >I currently have one computer (Compaq) running XP, and one with OS X and XP. >Both of them run as clean so far. These computers are exposed to a variety >of executable files as my whimsy dictates. > >I wish I understood more about how some people have continuous dire problems >that we manage to avoid. I think, in the case of Windows, it's mostly a matter of luck. We have a dozen identical HP machines, all running the identical (OEM!) XP/SP2 versions, and some are reliable, some are not. Some crash apps, some never crash those same apps. John |
| 09 May 2008 11:58:14 |
| John Larkin |
| Re: a dozen cpu's on a chip |
On Fri, 9 May 2008 18:51:23 +0100, "john jardine" <john.jardine@idnet.co.uk > wrote: > >"John Larkin" <jjlarkin@highNOTlandTHIStechnologyPART.com> wrote in message >news:kos824lnlilmk5pn9v5t698ga9li0f02bc@4ax.com... >[...] > >> [1] Could it be that programmers tend to look at programs as linearly >> executed screens of text, and realtime things - task switching, >> interrupts, async events - don't fit into their way of thinking? Maybe >> that's why the best programmers are people who were never educated as >> programmers. > > A programmer friend was yesterday showing me a shiny new book he'd just >bought to assist in some PC programming. >It's a big Yellow thing with 1400 pages dealing with C# and .NET. He then >took great amusement in randomly quoting inanities contained within. >Point is, as a 'self trained' hardware, C and assembler programmer, he >understood and could see the book for what it really is. >Very few 'computer science' graduates are capable of making this >distinction. > > At Borders Books, there's a huge computer section, but *nothing* about hardware-level stuff. There's a shrink-wrapped ".NET BASICS" set of books almost 2 feet long, must be ballpark 10,000 pages, for about $1200. We program in PowerBasic a lot, the Console Compiler version. You can write a 1-screen program to open a TCP/IP socket and do useful device control over Ethernet. For some reason, ex-chemists make very good programmers. John |
| 09 May 2008 21:28:48 |
| Martin Griffith |
| Re: a dozen cpu's on a chip |
On Wed, 07 May 2008 19:48:31 -0700, in sci.electronics.design John Larkin <jjlarkin@highNOTlandTHIStechnologyPART.com > wrote: > > >http://www.eetimes.com/news/latest/showArticle.jhtml;jsessionid=CESEXHWXUFJNKQSNDLSCKHA?articleID=207600531 > > >I bet we'll see 256 one of these days. > >John How about one of these http://royal.pingdom.com/?p=291 martin |
| 09 May 2008 13:15:53 |
| Joel Koltner |
| Re: a dozen cpu's on a chip |
"John Larkin" <jjlarkin@highNOTlandTHIStechnologyPART.com > wrote in message news:cd79241s44bebop8rdqgqa8958rjauh366@4ax.com... > At Borders Books, there's a huge computer section, but *nothing* about > hardware-level stuff. They're just reflecting demand, of course, although occasionally a place like Borders will have a "Getting started with PICs"-type of book If you go to a decent technical book store (such as Powell's Technical Book Store in Portland) there's tons of hardware books... there are also all sorts of intriging books on other nichey technical areas like soil analysis, dam constructions, ordinance disposal, etc. Walking through the aisles is a real eye-opener for just how "deep" some other fields are out there. The computer industry -- and Microsoft in particular -- seems to have realized that to continue to grow at the rate they'd like to, they have to constantly (attempt to) change the programming paradigm so that there's constant demand for continuous training, new tools, etc. Look at Vista: They don't officially support Visual Studio 6 (except for Visual BASIC), yet it's quite trivial to get C++ up and running as well (plenty of write-ups on the web, taking no more than a paragraph to describe the meat of the matter... and I suspect most of the other components of Visual Studio, but realistically >90% of Visual Studio users are probably looking for VC++ or VB6). It's hard to believe that Visual Studio 6 support was dropped for primarily technical rather than political reasons, i.e., Microsoft wants everyone to switch to C# and the other .Net languages. While good programmers can become noticeably more productive on, say, large applications by switching from, say, C to C++, a large percentage don't seem to make the "good" cut and hence end up writing VB.net code that's no better than the crappy old VB 6 code they used to write. The other big problem I see with programmers is that the not-so-good ones seem to want to make things far more complicated than the need to be... often with the full support of a big IT staff who needs to justify spending $25,000 on their next SQL Server box. See: http://thedailywtf.com/Articles/Two-Weeks-Notice.aspx ---Joel |
| 09 May 2008 16:44:38 |
| Michael A. Terrell |
| Re: a dozen cpu's on a chip |
gearhead wrote: > > The rule: for acronyms of five letters or more, capitalize only the > first letter > (e.g., Unicef). Don't waste your time trying to educate dimbulb. -- http://improve-usenet.org/index.html Use any search engine other than Google till they stop polluting USENET with porn and junk commercial SPAM If you have broadband, your ISP may have a NNTP news server included in your account: http://www.usenettools.net/ISP.htm |
| 09 May 2008 16:45:59 |
| Michael A. Terrell |
| Re: a dozen cpu's on a chip |
gearhead wrote: > > Yes, and if you break it Eeyore will spank you. This is the most pathetic post that I've ever had the misfortune to read. :( -- http://improve-usenet.org/index.html Use any search engine other than Google till they stop polluting USENET with porn and junk commercial SPAM If you have broadband, your ISP may have a NNTP news server included in your account: http://www.usenettools.net/ISP.htm |
| 09 May 2008 16:47:00 |
| Michael A. Terrell |
| Re: a dozen cpu's on a chip |
Joel Koltner wrote: > > Mine is closer to, "special rules for special people." That sounds like a politically correct definition of a kill filter. ;-) -- http://improve-usenet.org/index.html Use any search engine other than Google till they stop polluting USENET with porn and junk commercial SPAM If you have broadband, your ISP may have a NNTP news server included in your account: http://www.usenettools.net/ISP.htm |
| 09 May 2008 16:51:58 |
| Michael A. Terrell |
| Re: a dozen cpu's on a chip |
Jim Thompson wrote: > > On Fri, 9 May 2008 18:51:23 +0100, "john jardine" > <john.jardine@idnet.co.uk> wrote: > > > > >"John Larkin" <jjlarkin@highNOTlandTHIStechnologyPART.com> wrote in message > >news:kos824lnlilmk5pn9v5t698ga9li0f02bc@4ax.com... > >[...] > > > >> [1] Could it be that programmers tend to look at programs as linearly > >> executed screens of text, and realtime things - task switching, > >> interrupts, async events - don't fit into their way of thinking? Maybe > >> that's why the best programmers are people who were never educated as > >> programmers. > > > > A programmer friend was yesterday showing me a shiny new book he'd just > >bought to assist in some PC programming. > >It's a big Yellow thing with 1400 pages dealing with C# and .NET. He then > >took great amusement in randomly quoting inanities contained within. > >Point is, as a 'self trained' hardware, C and assembler programmer, he > >understood and could see the book for what it really is. > >Very few 'computer science' graduates are capable of making this > >distinction. > > > > > > From experience with two of my offspring, good programming seems to > come from good language skills (daughter and son are both > multi-lingual), irrespective of training in "computer science" (a term > I consider an oxymoron, right up there with "political science"). That's not a s bad as: "Another brilliant design by Bill Sloman". -- http://improve-usenet.org/index.html Use any search engine other than Google till they stop polluting USENET with porn and junk commercial SPAM If you have broadband, your ISP may have a NNTP news server included in your account: http://www.usenettools.net/ISP.htm |
| 09 May 2008 22:41:41 |
| Eeyore |
| Re: a dozen cpu's on a chip |
gearhead wrote: > Jim Thompson wrote: > > gearhead wrote: > > ><FatBytest...wrote: > > >> John Larkin wrote: > > > > >> >[1] I kept the apostrophe so that people whose expertise stops at 20 > > >> >KHz can have something clever to say. But to me, cpus looks awkward; > > >> >you'd pronounce it "seepuss." > > > > >> CPUs It's an acronym, so it should be capitalized. The rules on > > >> pluralizing an acronym are what need to be defined.- Hide quoted text - > > > > > > >The rule: for acronyms of five letters or more, capitalize only the > > >first letter (e.g., Unicef). > > > > Is there really a "rule" ?:-) > > > > ...Jim Thompson > > Yes, and if you break it Eeyore will spank you. I most certainly will ! ;~) Graham |
| 09 May 2008 19:12:14 |
| John Larkin |
| Re: a dozen cpu's on a chip |
On Fri, 09 May 2008 21:28:48 +0200, Martin Griffith <mart_in_medina@yah00.es > wrote: >On Wed, 07 May 2008 19:48:31 -0700, in sci.electronics.design John >Larkin <jjlarkin@highNOTlandTHIStechnologyPART.com> wrote: > >> >> >>http://www.eetimes.com/news/latest/showArticle.jhtml;jsessionid=CESEXHWXUFJNKQSNDLSCKHA?articleID=207600531 >> >> >>I bet we'll see 256 one of these days. >> >>John > > >How about one of these >http://royal.pingdom.com/?p=291 > > >martin In that third image down, those things towards the right sure look like floppy drives! John |
| 09 May 2008 22:27:56 |
| krw |
| Re: a dozen cpu's on a chip |
In article <3n6624pu6762nup9apu3crj5vh1uu6fqbn@4ax.com >, jjlarkin@highNOTlandTHIStechnologyPART.com says... > On Thu, 8 May 2008 07:42:04 -0700 (PDT), MooseFET <kensmith@rahul.net> > wrote: > > >On May 7, 7:48 pm, John Larkin > ><jjlar...@highNOTlandTHIStechnologyPART.com> wrote: > >> http://www.eetimes.com/news/latest/showArticle.jhtml;jsessionid=CESEX... > >> > >> I bet we'll see 256 one of these days. > > > >When you get to large numbers of CPUs it seems to make sense to stop > >making them identical. For servers this would be doubly so. Many of > >the CPUs won't need to do floating point operations. > > Right. Amybe a few cpu's would have serious floating point power, or a > few separate fp engines could be assigned to cpu's as needed. Lots of > cpu's, doing stuff like file i/o or serial stuff, could be less > powerful. I suppose we'll always need special graphics hardware, but > just a few of those per chip. Asymmetric multiprocessing makes the scheduler's life more complicated. Since the scheduler is part of the OS, and the OS is most often M$, this isn't a good idea, IMO. ;-) Hardware is cheap (so cheap PowerPC is including decimal FPUs). Throw the FPU on every node, whether its needed or not. > >It also would make sense to d |