similar to: R on Linux/Opteron?

Dear R-devel, Has anyone played with this? http://www.amd.com/us-en/Processors/DevelopWithAMD/0,,30_2252_2282,00.html <http://www.amd.com/us-en/Processors/DevelopWithAMD/0,,30_2252_2282,00.html> . I'll probably give it a shot... Best, Andy Andy Liaw, PhD Biometrics Research PO Box 2000, RY33-300 Merck Research Labs Rahway, NJ 07065 mailto:andy_liaw@merck.com

Before I do that, I would need to remove the gcj stuff that are in /usr/bin. If I know how to remove gcj, I'd gladly do that. However, for the particular version of the OS, the entire GCC seems to be bundled into one rpm, and I could not remove just the gcj component. Neither do I wish to mess with files that are part of some RPMs--- in my experience that's invitation for trouble later.

Dear all, Has anyone seen the same problem? I tried compiling R-1.9.0 alpha (2004-03-16) on our Opteron box running SUSE Linux ES8. I ran: ./configure --enable-R-shlib --with-blas=goto and got: Source directory: . Installation directory: /usr/local C compiler: gcc -m64 -O2 -g -msse2 -march=k8 -Wall --pedantic C++ compiler: g++ -m64 -O2 -g

Hi all, I've only now noticed that on our Opteron boxes running Suse Enterprise Server 8 that R-2.0.1 did not pass make chek-all. It bombed in tests/lapack.R at: > eigenok <- function(A, E, Eps = 1000 * .Machine$double.eps) { V <- E$vect lam <- E$values stopifnot(abs(A %*% V - V %*% diag(lam)) < Eps, .... [TRUNCATED] > Ceigenok <- function(A, E, Eps =

> From: Douglas Bates [mailto:bates at bates4.stat.wisc.edu] > > "Liaw, Andy" <andy_liaw at merck.com> writes: > > > Sorry. I need to retract my claim. There seems to be a 3G > limit, even > > though the OS could handle nearly 8G. (I can have two > simultaneous R > > processes each using near 3G.) > > > > On another note, on

> From: Douglas Bates [mailto:bates at bates4.stat.wisc.edu] > > "Liaw, Andy" <andy_liaw at merck.com> writes: > > > Sorry. I need to retract my claim. There seems to be a 3G > limit, even > > though the OS could handle nearly 8G. (I can have two > simultaneous R > > processes each using near 3G.) > > > > On another note, on

Dear R-help, Has anyone tried running R on a Beowulf-type cluster? I can get R to run in batch (using R CMD BATCH) on a cluster, but am wondering if it is possible to get an interactive R session on a compute node. Right now, if I run: beorun --nolocal R I just get the R start-up message and back to the shell prompt. If I try bpsh 0 R I can get R started (but the R prompt does not

Dear R-help, This one is probably a piece of cake for regexp masters. I'd like to split a character vector (for simplicity, say of length one for now) that contains fields that are delimited by arbitrary number of white spaces (e.g., " a b c "). How do I get the character vector that contain the fields? In the example I gave, I've tried: > strsplit(" a b c

Dear R-help, I'm trying to build R-2.4.0 on our Opteron-based Scyld cluster. The system has gcj (the GNU Java compiler, part of GCC) stuff in /usr/bin. When I installed jdk 1.5.08, the install script placed it in /usr/java (I didn't have a choice, as the script didn't offer that option). Now when I run configure in R-2.4.0, it finds gcj, which is not what I want to use. Is there a

Dear DevelopeRs, I'd like to ask those who develop R packages with compiled code to please try avoiding dependency on GCC (gcc/g77/gfortran/g++) specific features in the code, for the simple reason that there are non-GCC compilers out there that might choke on such features. I found this out back when I was trying to build R-2.3.0-to-be on our dual Opteron based Scyld cluster. Because the

Hi All, While attempting to build R-patched using the Portland Group compiler suite on our dual Opteron 250 box running Scyld 29cz5 (based on RH, kernel 2.4.29), I get: make[5]: Leaving directory `/home/andy/Rbuild/R221-pgi/src/library/tools/src' make[5]: Entering directory `/home/andy/Rbuild/R221-pgi/src/library/tools/src' pgcc -I../../../../include -I/include -I/include/CC -fpic

Dear R-help, Has anyone notice the problem that, on Windows (NT and XP), when printing a graph using the "File -> Print..." menu in the graphics window to print the graph, that line width seemed to be ignored in the printed output? For example, if I make a plot with plot(1:10, type="l", lwd=5), it looks right on screen, but when printed out using the menu, it looks like

Dear R-devel, [I'm not sure if this is appropriate for R-devel. If not, I'm more than happy to move it to R-help.] As those of you who saw my post on R-help know, I've been trying to build R-patched on a dual Opteron box running Scyld Beowulf, using the PGI 6.1 compilers. The build went fine, but I couldn't get it to pass make check-all. Jennifer Lai, who reported success with

Dear R-help, I was bitten by the behavior of all() when given logical(0): It is TRUE! (And any(logical(0)) is FALSE.) Wouldn't it be better to return logical(0) in both cases? The problem surfaced because some un-named individual called randomForest(x, y, xtest, ytest,...), and gave y as a two-level factor, but ytest as just numeric vector. I thought I check for that in my code by testing

Dear R-devel, Can anyone give me some hints on how to go about debugging a strange segfault in my randomForest package? Here's the scoop: A user reported segfault when running predict() in the randomForest package. I asked for the data and code. The combination runs fine under WinXPPro, but does give segfault on one of our Linux boxes running R (1.7.0 through R-devel_2004-01-08) on

Dear R-devel, I recall that there used to be a mechanism to get around the requirement that all objects in a package have associated documentation; i.e., a way to specify a list of objects (mostly functions) that are not considered as part of the package API. Is this still available? I cannot find any mention in R-exts. I realize that the way to go is to have a name space, and eventually

[I believe few users manipulate formulae and model frames like this, so I'm posting here instead of R-help.] I just found out that model.frame() includes all terms, including those preceded by "-" in the formula, in its output (as mentioned on its help page). Is there a recommended way of excluding "negative" terms? My guess is that I can use info in the terms attribute

Hi Martin, When I attended the LinuxWorld Expo in NYC back in January, I chatted with some folks at the AMD booth, as well as guys from Penguin Computing (where we bought our Opteron box). I was told that the Operton has this somewhat strange setup that the memory is controlled by one CPU. The net effect of this being that when both CPUs are running, one might only be running at around 90%

Dear R-help, I've been compiling R from source on our Linux boxes for quite a while. One thing that bugs me is that I always get an error when make check-all where it choked up running tests/internet.R. That wasn't a big deal, as I can run install.packages()/update.packages() fine, and that's all I use the 'net connection for. The problem is that it takes a _long_ time to fail

Dear R-devel, Apologies for bothering y'all with this seemingly perennial question. A user reported problem with my most recent version of randomForest (4.4-1), and I was able to reproduce it with his data with R-2.0.0 patched (2004-10-24) on WinXP Pro. The problem is that it crashes R on Windows. However, it does not happen on Linux (tried SUSE ES8 on our Opterons and Quantian on my