hkaiser changed the topic of #ste||ar to: STE||AR: Systems Technology, Emergent Parallelism, and Algorithm Research | stellar.cct.lsu.edu | HPX: A cure for performance impaired parallel applications | github.com/STEllAR-GROUP/hpx | Buildbot: http://rostam.cct.lsu.edu/ | Log: http://irclog.cct.lsu.edu/ | GSoD: https://developers.google.com/season-of-docs/
nikunj has joined #ste||ar
K-ballo has quit [Quit: K-ballo]
hkaiser has quit [Quit: bye]
JClave has joined #ste||ar
<JClave> Hi all, anyone here?
<zao> Yup.
nikunj has quit [Remote host closed the connection]
K-ballo has joined #ste||ar
hkaiser has joined #ste||ar
nikunj has joined #ste||ar
<tarzeau> i've got these http://phd-sid.ethz.ch/debian/hpx/problems.txt trying to package hpx 1.3.0 for debian, hints/help welcome
<tarzeau> and is something wrong when i run hpxcxx, it's telling me: "Cannot locate HPX"?
<nikunj> hkaiser: yt?
<hkaiser> nikunj: here
<nikunj> hkaiser: see pm please
<hkaiser> tarzeau: what do those problems mean?
<tarzeau> there's no manual pages for hpxcxx and hpxrun and the four hpx_* python scripts
<hkaiser> tarzeau: and I don't know anything about hpxcxx
<tarzeau> what do people do with hpx, just use hpxrun ? i should probably have a close look at the redhat/fedora hpx packaging, the freebsd one and the spack one
<hkaiser> tarzeau: I meant, the file you linked doesn't really explain what the problems are
<tarzeau> true, they're probably too much debian specific, except the rpath defines. i've only seen cmake configuration allows relative and absolute rpaths
<hkaiser> tarzeau: hpxrun is just a helper script simplifying distributed runs, it's not absolutely necessary, we use it for our testing infrastructure, however
<tarzeau> so basically hpx is just a library your own programs link against
<hkaiser> yes
<tarzeau> and it can be used with slurm together
<hkaiser> yes
<hkaiser> it picks up environment setting if run using slurm (same for other batch systems, btw)
<tarzeau> is condorht supported too?
<hkaiser> no
<hkaiser> hpx is currently targetting tightly coupled distributed applications (similar to MPI style apps)
<hkaiser> tarzeau: gtg now, sorry - let's talk later
<tarzeau> tomorrow
<tarzeau> thanks
<hkaiser> tarzeau: heller or simbergm can help for sure
quaz0r has quit [Ping timeout: 245 seconds]
quaz0r has joined #ste||ar
JClave has quit [Quit: Going offline, see ya! (www.adiirc.com)]
<heller> tarzeau: we don't have condor integration, do you have a usecase for that? should be perfectly doable though
<heller> tarzeau: hpxcc is just a wrapper script that calls pkgcfg to add the necessary flags etc.
<heller> the other hpx_ python scripts should probably be not installed in the first place
<heller> hpxrun is a very useful utility to run HPX distributed on a single node for testing purposes. We extended it to slurm to ease our testing setup
<heller> hkaiser: appear.in?
<hkaiser> heller: ok, sec
jaafar has quit [Ping timeout: 252 seconds]
<tarzeau> heller: not at all, we've moved to slurm (had condor 10 years ago, then torque)
<tarzeau> heller: so i'll keep hpxcc/hpxrun but drop the python scripts (i also turned off building examples, what is your opinion on that?)
<hkaiser> tarzeau: no reason not to drop the examples, they are not needed
<hkaiser> they are just that - examples..
jaafar has joined #ste||ar
jaafar has quit [Ping timeout: 245 seconds]
jaafar has joined #ste||ar
diehlpk_mobile has joined #ste||ar
<hkaiser> diehlpk_mobile: panel was rejected :(
<diehlpk_mobile> see im
jaafar has quit [Ping timeout: 245 seconds]
Coldblackice has quit [Ping timeout: 258 seconds]