How Many Processor Cores Are Enough?
|
From: Eugene Miya on 4 Oct 2006 20:28 In article <877izgrq0x.fsf(a)k9.prep.synonet.com>, <prep(a)prep.synonet.com> wrote: >eugene(a)cse.ucsc.edu (Eugene Miya) writes: >> I saved a DAP for the CHM, and I used the BBN and I think I have >> succeeded to locating one surviving representative sitting out in a >> field near Denver. > >It would be nice to think you are joking, but having done that stuff >myself... Like getting a DEC 630(?) display out of a `off to the tip' >pile. Naw finding old machines is part of my job. >Do you know of a working 7 track tape drive? We have some data from >Apollo 12, and it MAY be 7 track. I believe that an old boss of mine (she died) lucked out and found a set of heads and had Kennedy (near JPL, not KSC) built a tape drive as a project. It's just finding it. --
From: Chris Thomasson on 4 Oct 2006 21:21 "Eugene Miya" <eugene(a)cse.ucsc.edu> wrote in message news:45215a91$1(a)darkstar... >>>> "Nick Maclaren" <nmm1(a)cus.cam.ac.uk> wrote in message >>>>> In article <451d8b01(a)darkstar>, eugene(a)cse.ucsc.edu (Eugene Miya) [...] > In article <Up-dnUuwdY8SoILYnZ2dnUVZ_rCdnZ2d(a)comcast.com>, > Chris Thomasson <cristom(a)comcast.net> wrote: >>"Bill Todd" <billtodd(a)metrocast.net> wrote in message >>news:VPWdnbJFjNGrvoLYnZ2dnUVZ_uudnZ2d(a)metrocastcablevision.com... >>> Chris Thomasson wrote: >>> Hmmm - you appear to be about as delusional in that area [...] >>> as you are about the relative importance of lock-free algorithms. [...] >>What is your take on the issue? Perhaps you have misunderstood me; a >>majority of my postings are directed to comp.programming.threads). FWIW, I >>do believe that a clever mixture of lock-free, and lock-based algorithms >>can >>produce results that can scale extremely well. For instance, here is an >>older library of mine: > ... >>"An Effective Marriage between Lock-Free and Lock-Based Algorithms" > > Is this about Henry M.'s lock free OS of old? This one correct: http://citeseer.ist.psu.edu/update/50389 I just briefly skimmed through the paper... Some of the lock-free techniques in that OS depend on being executed in kernel mode... They make use of DWCAS which is not portable at all... I need to read the whole paper, then I can better comment... Anyway, my AppCore library is about brining advanced lock-free techniques out of the operating system, and into user-space. It also shows how to mix lock-free and lock-based techniques together, to get the best of both worlds... I also wanted to provide the public with an example of a full-blown working library that has lock-free techniques scattered throughout its infrastructure... Apparently, it kind of "seems" to be catching on a little bit: http://groups.google.com/group/comp.programming.threads/browse_frm/thread/205dcaed77941352 ;)
From: nobody on 5 Oct 2006 19:09 >Today I read that we're going to get quad-core processors >in 2007, and 80-core processors in 5 years. Azul Systems is shipping boxes with 24 core chips for 18 months. How many are enough depends on how many threads you have.
From: ranjit_mathews@yahoo.com on 9 Oct 2006 15:13 Terje Mathisen wrote: > ranjit_mathews(a)yahoo.com wrote: > > How does the Itanium compare in terms of SpecInts PER GHz? > > > Spec/GHz is very nearly totally meaningless. Then why does the Core 2 Duo find favor over Pressler core processors? > Why do you care? > > Terje > > -- > - <Terje.Mathisen(a)hda.hydro.com> > "almost all programming can be viewed as an exercise in caching"
From: Greg Lindahl on 9 Oct 2006 17:07
In article <1160421201.367674.89490(a)c28g2000cwb.googlegroups.com>, ranjit_mathews(a)yahoo.com <ranjit_mathews(a)yahoo.com> wrote: >> Spec/GHz is very nearly totally meaningless. > >Then why does the Core 2 Duo find favor over Pressler core processors? Because Core2 is faster at lower power? Note that "GHz" doesn't appear in the sentence; Ghz is just an implementation detail, not comparable between designs. But "faster" and "lower power", those are comparable. -- greg |