search for: ncpus

I have a Ubuntu Linux 12.04.1 machine running R 2.15.1. I'm trying to run tests building packages, so install.packages() is trying to install about about 1000 packages from source. I'm running into some strange issues that seem related to using values for Ncpus other than 1. When I use Ncpus=32, about 260 of the packages fail to install. When I tried again with Ncpus=2, more of them get installed, and there were only 234 failed packages. I tried it again with Ncpus=1, after which there were only 142 failed packages. (I think these remaining failed packag...

...4 cpus and most of the packages are written in >> C/C++. >> >> So is it possible to add a -j4 flag to the make command when I use the >> install.packages() function? >> That will probably speed up the package installation process 390%. >> > > See the Ncpus argument in ?install.packages. Thanks. However it looks like Ncpus=4 tries to compile four R packages at the same time using one cpu for each packages. From the documentation: " Ncpus: the number of parallel processes to use for a parallel install of more than one source packa...

...i); + wrmsr(HV_X64_MSR_SINT0 + i, 0xFF|HV_SYNIC_SINT_MASKED); + } + + wrmsr(HV_X64_MSR_SCONTROL, 0); + wrmsr(HV_X64_MSR_SIMP, 0); + wrmsr(HV_X64_MSR_SIEFP, 0); + atomic_inc(&cpus_comp_count); +} + +int main(int ac, char **av) +{ + + if (synic_supported()) { + int ncpus, i; + + setup_vm(); + smp_init(); + setup_idt(); + enable_apic(); + + synic_prepare_sint_vecs(); + + ncpus = cpu_count(); + if (ncpus > MAX_CPUS) { + ncpus = MAX_CPUS; + } + printf("ncpus = %d\n", ncpus); + +...

...i); + wrmsr(HV_X64_MSR_SINT0 + i, 0xFF|HV_SYNIC_SINT_MASKED); + } + + wrmsr(HV_X64_MSR_SCONTROL, 0); + wrmsr(HV_X64_MSR_SIMP, 0); + wrmsr(HV_X64_MSR_SIEFP, 0); + atomic_inc(&cpus_comp_count); +} + +int main(int ac, char **av) +{ + + if (synic_supported()) { + int ncpus, i; + + setup_vm(); + smp_init(); + setup_idt(); + enable_apic(); + + synic_prepare_sint_vecs(); + + ncpus = cpu_count(); + if (ncpus > MAX_CPUS) { + ncpus = MAX_CPUS; + } + printf("ncpus = %d\n", ncpus); + +...

...D); > + } > + > + wrmsr(HV_X64_MSR_SCONTROL, 0); > + wrmsr(HV_X64_MSR_SIMP, 0); > + wrmsr(HV_X64_MSR_SIEFP, 0); > + atomic_inc(&cpus_comp_count); > +} > + > +int main(int ac, char **av) > +{ > + > + if (synic_supported()) { > + int ncpus, i; > + > + setup_vm(); > + smp_init(); > + setup_idt(); > + enable_apic(); > + > + synic_prepare_sint_vecs(); > + > + ncpus = cpu_count(); > + if (ncpus > MAX_CPUS) { > + ncpus = MAX_CPUS; > +...

...re written in >>> C/C++. >>> >>> So is it possible to add a -j4 flag to the make command when I use the >>> install.packages() function? >>> That will probably speed up the package installation process 390%. >>> >> >> See the Ncpus argument in ?install.packages. > > > Thanks. > > However it looks like Ncpus=4 tries to compile four R packages at the same > time using one cpu for each packages. > But you said you had lots to install so would that not speed things up too? > From the documentation:...

...internal, available = NULL, destdir = ${DEST}, dependencies = NA, type = getOption("pkgType"), configure.args = getOption("configure.args"), configure.vars = getOption("configure.vars"), clean = FALSE, Ncpus = getOption("2", 1L), libs_only = FALSE, INSTALL_opts,--verbose) and this is the result stats [ ~ ]# export DIR=/home/stats/R-2.15.0_runTEST190712A stats [ ~ ]# export DEST=/home/stats/Rtester stats [ ~ ]# cd $DEST stats [ ~/Rtester ]# stats [ ~/Rtester ]# ${DIR}/bin/R...

Hi, I am installing a lot of packages to a new R installation and it takes a long time. However the machine got 4 cpus and most of the packages are written in C/C++. So is it possible to add a -j4 flag to the make command when I use the install.packages() function? That will probably speed up the package installation process 390%. Regards Martin M. S. Pedersen [[alternative HTML

...time. > However the machine got 4 cpus and most of the packages are written in > C/C++. > > So is it possible to add a -j4 flag to the make command when I use the > install.packages() function? > That will probably speed up the package installation process 390%. See the Ncpus argument in ?install.packages. Duncan Murdoch

...mmin2068 wrote: > I will say one thing that I noticed is: when giving the guest 8 cores > and not setting the topology, windows device manager shows 8 CPUs but > the task manager only shows 2. Feh. Windows. Historically, if you did NOT specify topology then QEMU would defaults to sockets==nCPUs, cores==1, threads=1. Since QEMU 6.2, new machine types for x86 will prefer cores, IOW, if you don't specify topology then QEMU would give you sockets=1, cores=nCPUs, threads=1 The rationale for this is that it matches common physical hardware. It is rare to see an x86 machine with 32 sockets...

...unsigned credits_per_tslice; }; static void csched_tick(void *_cpu); @@ -326,7 +325,7 @@ csched_free_pdata(const struct scheduler spin_lock_irqsave(&prv->lock, flags); - prv->credit -= CSCHED_CREDITS_PER_ACCT; + prv->credit -= prv->credits_per_tslice; prv->ncpus--; cpu_clear(cpu, prv->idlers); cpu_clear(cpu, prv->cpus); @@ -360,19 +359,19 @@ csched_alloc_pdata(const struct schedule spin_lock_irqsave(&prv->lock, flags); /* Initialize/update system-wide config */ - prv->credit += CSCHED_CREDITS_PER_ACCT; + prv->...

...compile job, which is again a facility that ninja has. > If this is a fact, can you please adjust the information in [1]? > > Do I have other options to speedup my build? Yes: start to use ninja ;) If for any reason (?) you are really stuck with "make", then run "make -j ~ncpus" (with ~ncpus "approximately the number of cores in your machine). It will build in parallel. -- Mehdi > ( For testing RCs I tend to use slow build - one single (c)make job. ) > > My build-script and configure-log are attached. > > Thanks. > > Regards, > -...

Dear R-devel list, R's detectCores() function internally calls "ncpus" function to get the total number of logical processors. However, this doesnot seem to take NUMA into account on Windows machines. On a machine having 48 processors (24 cores) in total and windows server 2012 installed, if NUMA is enabled and has 2 nodes (node 0 and node 1 each having 24 CPUs...

...the configure stage. Configure seems to be looking for zlib and fails even when I specify --without-zlib or --disable-zlib. I'm enclosing the build log as part of this bug report. Regards and THANKS for the cool software. George... cat bld.log1 #!/bin/sh -xv if [ $# = 0 ] then ncpus=1 else ncpus=$1 fi + [ 0 = 0 ] ncpus=1 sysver=`uname -s` + uname -s sysver=SunOS syslvl=`uname -r` + uname -r syslvl=5.8 osname=$sysver.$syslvl osname=SunOS.5.8 case $osname in SunOS.5.5) osname="$sysver.5.5" ;; SunOS.5.5.1) osname="$sysve...

...t;> >>> If this is a fact, can you please adjust the information in [1]? >>> >>> Do I have other options to speedup my build? >> >> Yes: start to use ninja ;) >> If for any reason (?) you are really stuck with "make", then run "make -j ~ncpus" (with ~ncpus "approximately the number of cores in your machine). It will build in parallel. >> > > So, I like NinjaS but I have not used it yet as software. > > From my build-script... > > [ -d ${BUILD_DIR} ] || mkdir -p ${BUILD_DIR} > cd $BUILD_DIR > >...

...facility that ninja has. > >> If this is a fact, can you please adjust the information in [1]? >> >> Do I have other options to speedup my build? > > Yes: start to use ninja ;) > If for any reason (?) you are really stuck with "make", then run "make -j ~ncpus" (with ~ncpus "approximately the number of cores in your machine). It will build in parallel. > So, I like NinjaS but I have not used it yet as software. >From my build-script... [ -d ${BUILD_DIR} ] || mkdir -p ${BUILD_DIR} cd $BUILD_DIR # BUILD-VARIANT #1: CONFIGURE AND MAKE (w...

...method, available = NULL, destdir = NULL, dependencies = NA, type = getOption("pkgType"), configure.args = getOption("configure.args"), configure.vars = getOption("configure.vars"), clean = FALSE, Ncpus = getOption("Ncpus", 1L), verbose = getOption("verbose"), libs_only = FALSE, INSTALL_opts, quiet = FALSE, keep_outputs = FALSE, ...) ... Arguments to be passed to download.file or to the functions for binary installs on OS X an...

...[0-9]+[ \t]*", out, value=TRUE)))) # number of physical processors - as a sanity check system("wmic computersystem get numberofprocessors") Thanks, Tomas On 08/17/2018 05:11 PM, Srinivasan, Arunkumar wrote: Dear R-devel list, R's detectCores() function internally calls "ncpus" function to get the total number of logical processors. However, this doesnot seem to take NUMA into account on Windows machines. On a machine having 48 processors (24 cores) in total and windows server 2012 installed, if NUMA is enabled and has 2 nodes (node 0 and node 1 each having 24 CPUs...

...ot;, out, value=TRUE)))) # number of physical processors - as a sanity check system("wmic computersystem get numberofprocessors") Thanks, Tomas On 08/17/2018 05:11 PM, Srinivasan, Arunkumar wrote: > Dear R-devel list, > > R's detectCores() function internally calls "ncpus" function to get the total number of logical processors. However, this doesnot seem to take NUMA into account on Windows machines. > > On a machine having 48 processors (24 cores) in total and windows server 2012 installed, if NUMA is enabled and has 2 nodes (node 0 and node 1 each havin...

Hi R-users, I have a problem running my R code on a Linux cluster. What I did was write a .pbs file to instruct the cluster on what to do and how: #!/bin/sh #PBS -m ae #PBS -M uqlcatta@uq.edu.au #PBS -A uq-CSER #PBS -N job1_lollo #PBS -l select=1:ncpus=1:NodeType=fast:mem=8GB #PBS -l walltime=999:00:00 cd $PBS_O_WORKDIR source /usr/share/modules/init/bash module load R/2.11.1 /home/uqlcatta/script/diag.sh The .pbs file calls a .sh file, which is located on my home directory on the cluster, and which contains the R script (...