search for: cohn

...released in November. KNL is not a co-processor so no offload is necessary. R executes directly on the Phi, which looks like a multi-core machine with 64 cores. Robert -----Original Message----- From: Tomas Kalibera [mailto:tomas.kalibera at gmail.com] Sent: Monday, January 16, 2017 12:00 PM To: Cohn, Robert S <robert.s.cohn at intel.com> Cc: r-devel at r-project.org Subject: Re: [Rd] accelerating matrix multiply Hi Robert, thanks for the report and your suggestions how to make the NaN checks faster. Based on my experiments it seems that the "break" in the loop actually can...

...if(2),nrow=m,ncol=n) print(typeof(A[1,2])) print(A[1,2]) # Matrix multiply system.time (C <- B %*% A) system.time (C <- B %*% A) system.time (C <- B %*% A) -----Original Message----- From: Martin Maechler [mailto:maechler at stat.math.ethz.ch] Sent: Tuesday, January 10, 2017 8:59 AM To: Cohn, Robert S <robert.s.cohn at intel.com> Cc: r-devel at r-project.org Subject: Re: [Rd] accelerating matrix multiply >>>>> Cohn, Robert S <robert.s.cohn at intel.com> >>>>> on Sat, 7 Jan 2017 16:41:42 +0000 writes: > I am using R to multiply some lar...

...turn TRUE; return FALSE; } And this was the "nobreak" version: static __attribute__ ((noinline)) Rboolean hasNA(double *x, int n) { Rboolean has = FALSE; for (R_xlen_t i = 0; i < n; i++) if (ISNAN(x[i])) has=TRUE; return has; } Thanks, Tomas On 01/11/2017 02:28 PM, Cohn, Robert S wrote: >> Do you have R code (including set.seed(.) if relevant) to show on how to generate >> the large square matrices you've mentioned in the beginning? So we get to some >> reproducible benchmarks? > > Hi Martin, > > Here is the program I used. I onl...

I am using R to multiply some large (30k x 30k double) matrices on a 64 core machine (xeon phi). I added some timers to src/main/array.c to see where the time is going. All of the time is being spent in the matprod function, most of that time is spent in dgemm. 15 seconds is in matprod in some code that is checking if there are NaNs. > system.time (C <- B %*% A) nancheck: wall time

I think the thing that is missing is a simple way for end users on windows to replace blas/lapack libraries with MKL-a package that you install that puts the libraries in the right place. Microsoft provides something for their distro, but we don't have the equivalent if you get R from cran. On 29 October 2017 at 22:01, Kenny Bell wrote: | User here: incorporating Intel's MKL, as MRO

We have just had an ISDN PRI service commissioned here in AU (using Powertel as provider). I have called them and ensured that we have the ability to set Caller ID on our service to any number in our 100 number block, and I have been assured that everything is OK from their end (unlikely). Every time I drop a sample.call in to /var/spool/asterisk/outgoing with the Callerid: option set, the

Is anyone successfully using this phone with Asterisk? There is a lot mentioned about CISCO but nothing about Nortel... Alex.

> From: "Cohn, Robert S" <robert.s.cohn at intel.com> > > I am using R to multiply some large (30k x 30k double) matrices on a > 64 core machine (xeon phi). I added some timers to > src/main/array.c to see where the time is going. All of the time is > being spent in the matprod func...

> I would appreciate any insights over compiling R 3.4 with Intel MKL -- I have been successful until R 3.3.3 but now it stops complaining about pcre though it worked without Intel MKL as follows, I successfully built R-rc_2017-04-19_r72555.tar.gz with icc & MKL on centos 7 with this: # https://software.intel.com/en-us/articles/build-r-301-with-intel-c-compiler-and-intel-mkl-on-linux#

>>>>> Cohn, Robert S <robert.s.cohn at intel.com> >>>>> on Sat, 7 Jan 2017 16:41:42 +0000 writes: > I am using R to multiply some large (30k x 30k double) > matrices on a 64 core machine (xeon phi). I added some timers > to src/main/array.c to see where the time is going. A...

...ird option? Thanks in advance for you insights, -- There are no Debian developers in any part of Hell, because the good karma incurred by being one takes you straight to the pearly gates. Of course, the frequent flame wars you put up with on the Debian lists make up for this on Earth. - Seth Cohn

Hi, On Fri, Jun 4, 2010 at 5:18 AM, Pekka Nikander <pekka.nikander at nomadiclab.com> wrote: >  Would the best way be to add an option to -loop-unroll, and hack away at lib/Transforms/Utils/LoopUnroll.cpp? Instead, the better alternative is to write another pass similar to LoopUnrollPass.cpp (say LoopPeelPass.cpp) and add new option -loop-peel. The new pass could use llvm::UnrollLoop()

Full_Name: Tim Cohn Version: 1.6.1 OS: Macintosh OS X Submission from: (NULL) (130.11.34.250) The Mac version of survreg does not handle left-censored data correctly (at least the results are not what I get doing it other ways, and they are not the same as I get running R 1.6.1 in Windows 98se; the Windows 98 resul...

...0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 (Dispersion parameter for gaussian family taken to be 6934.706) Null deviance: 70776718 on 31 degrees of freedom Residual deviance: 180302 on 26 degrees of freedom AIC: 377.19 Convergence attained in 2 iterations with relative change -1.662839e-14 -- Greg Cohn Forestry Technician USDA Forest Service Fire Lab [[alternative HTML version deleted]]

...d up the call. Is this a known issue? Thanks, -- There are no Debian developers in any part of Hell, because the good karma incurred by being one takes you straight to the pearly gates. Of course, the frequent flame wars you put up with on the Debian lists make up for this on Earth. - Seth Cohn

Folks, Thanks to Jerry Jones, David Josephson, John Novack, and George Cohn for their posts about how a key system phone works. Now I'm starting to scrounge around for a KSU so I can look into the Collector's Net work :-) Most of the phones in the house are (fairly) new single-line analog jobs; but, since "retro-telephony" is an amusing part of this hob...

On Wed, 13 Nov 2002, Jan de Leeuw wrote: > > No problemo. And, in fact, I get the same results in > the R-1.6.0 Carbon version. I don't. Could there be a G3/G4 issue? -thomas > --- Jan > > On Wednesday, November 13, 2002, at 02:05 PM, tim@timcohn.com wrote: > > > Full_Name: Tim Cohn > > Version: 1.6.1 > > OS: Macintosh OS X > > Submission from: (NULL) (130.11.34.250) > > > > > > The Mac version of survreg does not handle left-censored data > > correctly (at > > least the results are n...

I am working on heavily optimising unusually static C++ code, and have encountered a situation where I basically want an optimiser that would speculatively unroll a loop to see if the first round of the loop could be optimised further. (I happen to know that it is possible.) The previous optimisations that produce the loop in the first place already do a magical job (relying heavily on constant

...No problemo. And, in fact, I get the same results in >>> the R-1.6.0 Carbon version. >> >> I don't. Could there be a G3/G4 issue? >> >> -thomas >> >> >>> --- Jan >>> >>> On Wednesday, November 13, 2002, at 02:05 PM, tim@timcohn.com wrote: >>> >>>> Full_Name: Tim Cohn >>>> Version: 1.6.1 >>>> OS: Macintosh OS X >>>> Submission from: (NULL) (130.11.34.250) >>>> >>>> >>>> The Mac version of survreg does not handle left-censored data...

...hms and Architectures for Parallel Processing, volume 6081 of Lecture Notes in Computer Science, pages 499{510. Springer Berlin / Heidelberg, 2010. [7] Maxwell Souza, Daniel Nic acio, and Guido Ara ujo. ISAMAP: Instruction Mapping Driven by Dynamic Binary Translation. In Mauricio Breternitz, Robert Cohn, Erik Altman, and Youfeng Wu, editors, AMAS-BT - 3rd Workshop on Architectural and Microarchitectural Support for Binary Trans- lation, Saint Malo France, 2010. [8] D. Ung and C. Cifuentes. Optimising hot paths in a dynamic binary trans- lator. In Workshop on Binary Translation, October 2000. -----...