hkaiser changed the topic of #ste||ar to: STE||AR: Systems Technology, Emergent Parallelism, and Algorithm Research | stellar-group.org | HPX: A cure for performance impaired parallel applications | github.com/STEllAR-GROUP/hpx | This channel is logged: irclog.cct.lsu.edu
<hkaiser> gnikunj[m]: as expected ;-)
<PatrickDiehl[m]> gonidelis What do you want to do with APEX?
<gonidelis[m]> Patrick Diehl: find hotspots on an hpx application i wrote
<PatrickDiehl[m]> Compile with APEX is the first thing to do
<PatrickDiehl[m]> You can either use OTF2, but you need to compile it
<PatrickDiehl[m]> Or Google trace
<PatrickDiehl[m]> The latter one can be read with Google chrome
<PatrickDiehl[m]> OTF2 get really large
<PatrickDiehl[m]> Vtunes is limited in the amount of threads
<PatrickDiehl[m]> Or traveler could be some option, however, I never used it
<PatrickDiehl[m]> https://t.co/SXxPYa0Rse
diehlpk has joined #ste||ar
diehlpk has quit [Quit: Leaving.]
K-ballo has quit [Quit: K-ballo]
diehlpk has joined #ste||ar
hkaiser has quit [Quit: Bye!]
diehlpk has quit [Quit: Leaving.]
Yorlik has joined #ste||ar
<Yorlik> Do we have lock free concurrent datastructures in HPX we can use, like lock free c-vector or lock free c-hashmap?
Yorlik has quit [Ping timeout: 264 seconds]
Yorlik has joined #ste||ar
K-ballo has joined #ste||ar
hkaiser has joined #ste||ar
Yorlik has quit [Ping timeout: 264 seconds]
hkaiser has quit [Quit: Bye!]
diehlpk has joined #ste||ar
diehlpk has quit [Quit: Leaving.]
diehlpk has joined #ste||ar
Yorlik has joined #ste||ar
hkaiser has joined #ste||ar
diehlpk_work has joined #ste||ar
hkaiser_ has joined #ste||ar
hkaiser has quit [Ping timeout: 264 seconds]
hkaiser_ has quit [Ping timeout: 264 seconds]
diehlpk_work has quit [Ping timeout: 264 seconds]
hkaiser has joined #ste||ar
<Yorlik> hkaiser: YT?
diehlpk_work has joined #ste||ar
<gonidelis[m]> is there some straightforward way to conver a c++ for loop to hpx::for_loop without haveing to deal with iterator initialization or sth? i mean for(int x = sth; x < sth_else; x++) -> how could i model sth and sth elese within the for_loop hpx algo?
<gnikunj[m]> gonidelis: try boost irange
hkaiser has quit [Quit: Bye!]
<gonidelis[m]> hhmmm do we have that in hpx?
<gnikunj[m]> nope
diehlpk has quit [Quit: Leaving.]
diehlpk has joined #ste||ar
<gonidelis[m]> fuk
diehlpk has quit [Quit: Leaving.]
hkaiser has joined #ste||ar
diehlpk has joined #ste||ar
diehlpk has quit [Quit: Leaving.]
<Yorlik> hkaiser: YT?
<Yorlik> Is there anything in HPX, that allows ordered message broadcasting, like zookeeper or similar systems?
<Yorlik> Actually I'm not even sure I need it. It's a tricky database question. Because I don't even fully understand if the system is a distributed database or not really. :D
<hkaiser> Yorlik: nothing ordered, you need to ensure ordering yourself
<Yorlik> Allright. So I just need to figure out if I really need it. But I'm afraid I do.
<Yorlik> I want to implement tagged object search where a tag basically is a set of objects with the same tag.
<hkaiser> nod
<hkaiser> pass the sequencing info as an argument
<Yorlik> Like a hash of node_id+time?.
<Yorlik> I'm currently reading about this lagorithm: https://gridem.blogspot.com/2016/05/replicated-object-part-7-masterless.html
<hkaiser> whatever works for you
<Yorlik> Seems I got some learning to do. Just prove to myself in an example that order does matter in my case with a simnple example.
<gnikunj[m]> hkaiser: ok, the code now works and executes properly. Let me finally convert it to resilient (should be trivial) and run some performance tests. I'll update you with the results.
<hkaiser> gnikunj[m]: \o/
<hkaiser> many thanks!
Yorlik has quit [Ping timeout: 264 seconds]
diehlpk has joined #ste||ar
diehlpk has quit [Quit: Leaving.]
K-ballo has quit [Quit: K-ballo]
K-ballo has joined #ste||ar