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/
hkaiser has joined #ste||ar
hkaiser has quit [*.net *.split]
diehlpk_work has quit [*.net *.split]
bibek has quit [*.net *.split]
simbergm has quit [*.net *.split]
tarzeau has quit [*.net *.split]
Yorlik has quit [*.net *.split]
quaz0r has quit [*.net *.split]
parsa has quit [*.net *.split]
jaafar_ has quit [*.net *.split]
david_pfander has quit [*.net *.split]
mdiers_ has quit [*.net *.split]
daissgr has quit [*.net *.split]
Vir has quit [*.net *.split]
Amy1 has quit [*.net *.split]
heller has quit [*.net *.split]
wash[m] has quit [*.net *.split]
zao has quit [*.net *.split]
jbjnr has quit [*.net *.split]
jgurhem has quit [*.net *.split]
K-ballo has quit [*.net *.split]
bobakk3r has quit [*.net *.split]
nikunj has quit [Remote host closed the connection]
simbergm has joined #ste||ar
bobakk3r has joined #ste||ar
K-ballo has joined #ste||ar
quaz0r has joined #ste||ar
bibek has joined #ste||ar
parsa has joined #ste||ar
diehlpk_work has joined #ste||ar
hkaiser has joined #ste||ar
jgurhem has joined #ste||ar
jbjnr has joined #ste||ar
tarzeau has joined #ste||ar
zao has joined #ste||ar
wash[m] has joined #ste||ar
heller has joined #ste||ar
Amy1 has joined #ste||ar
mdiers_ has joined #ste||ar
daissgr has joined #ste||ar
Vir has joined #ste||ar
simbergm has quit [Changing host]
simbergm has joined #ste||ar
simbergm has quit [Ping timeout: 240 seconds]
jaafar has joined #ste||ar
david_pfander has joined #ste||ar
Yorlik has joined #ste||ar
<Yorlik> hkaiser: yt?
<hkaiser> Yorlik: here
<Yorlik> Hello! Did you get my PM ? I got a netsplit shortly after sending.
<Yorlik> Basically coming back to the documentation talk we had some time ago.
<hkaiser> sorry, have not seen it
<hkaiser> but sure, let's talk about documentation at some point
<Yorlik> I was just asking when you had time to talk about it.
<Yorlik> When would you have time?
<hkaiser> we'll have a meeting with the new student from GSoD this Friday, that might be a good opportunity to get involved
<Yorlik> So - tomorrow?
<hkaiser> Friday
<hkaiser> 10am Central, I believe - that's 17:00
<hkaiser> for you
<Yorlik> it's thursday here already :)
<hkaiser> right
<hkaiser> parsa and simbergm organized that - talk to them ;-)
<Yorlik> The problem is I can't just start documenting stuff. We talked about me asking someone in the know and documenting that.
<hkaiser> I think diehlpk_work is involved as well
<hkaiser> yah sure
<hkaiser> do you have questions?
<Yorlik> Always.
<hkaiser> lol
<Yorlik> They pop up as I work.
<hkaiser> I'm too tired today, but we can try tomorrow
<hkaiser> can you collect those questions?
<Yorlik> But they also appear in a rough orser, since I have to go like 1-2-3 roughly.
<Yorlik> Sure.
<Yorlik> I'll poke you!
<hkaiser> k
hkaiser has quit [Quit: bye]
simbergm has joined #ste||ar
rori has joined #ste||ar
hkaiser has joined #ste||ar
<Yorlik> hkaiser: yt ?
<hkaiser> here
<Yorlik> Do you have time to talk?
<hkaiser> having my morning coffee right now ;-)
<hkaiser> will have to disapprear soon
<Yorlik> Allright. What time (roughly) might work?
<hkaiser> 10 my time?
<Yorlik> 10 a.m.?
<Yorlik> And what's your TZ again? Loisiana, right?
<hkaiser> that's 7 hours behind central Europe
<Yorlik> Sounds good / I|ll be there.
aserio has joined #ste||ar
hkaiser has quit [Quit: bye]
<diehlpk_work> Yorlik, The student has already a defined project to work on
<diehlpk_work> This it she project she will work on https://developers.google.com/season-of-docs/docs/participants/project-hpx
<Yorlik> I think the way how I could help is different. I need to get pestering hkaiser on a regular basis and document what he tells me. Mz idea is to make something like mz personal FAQ with example solutions.
<diehlpk_work> We have to keep in mind that she is a technical writer, she studies at LSU and has some work experience.
<diehlpk_work> So we can not ask her to document fancy C++ stuff
<diehlpk_work> She will improve the easier parts of the docs and we can use this as one example how to do it better
<Yorlik> I'm just a self taught guy in the middle of the C++ learning curve. So - I can provide the newbie perspective.
<Yorlik> Is there a way to have certain persistent objects on a worker thread which can be used by all tasks running on that worker thread, e.g. a LuaState?
<Yorlik> Our LuaStates will not create garbage or keep any state except global functions which are exposed after initialization.
<Yorlik> So we can cleanly use these functions by tasks. But tearing down and building up these states per task would be very inefficient.
hkaiser has joined #ste||ar
<Yorlik> Allright hkaiser - I'd be ready when you are.
<hkaiser> Yorlik: appear.in?
<Yorlik> Sure! What was the room id again?
<hkaiser> appear.in/stellar-group
nikunj has joined #ste||ar
<K-ballo> I just found out about the hpx intro course at cscs from the italian c++ slack
<aserio> K-ballo: You mean the CSCS tutorial?
<aserio> that's great
<aserio> How did that come up?
<K-ballo> nur 6:21 AM
<K-ballo> per chi fosse interessato in HPX ci sarà un corso a riguardo tenuto dagli sviluppatori John e Mikael a Lugano il 17-18 ottobre:
<aserio> K-ballo: Sweet, this is the upcoming one!
rori has quit [Ping timeout: 272 seconds]
aserio has quit [Ping timeout: 264 seconds]
aserio has joined #ste||ar
<Yorlik> hkaiser: I made an issue for the id_type persistence problem and also pasted the draft we made into it: https://github.com/STEllAR-GROUP/hpx/issues/4034
<Yorlik> Also heller: ^
<heller> simbergm: jbjnr: If you need some assistance, let me know
mreese3_ has joined #ste||ar
<mreese3_> Does anyone happen to know if the physl executable can parse the distributed data structures and operations?
<simbergm> K-ballo: nur is our colleague, he's eager at advertising hpx
<simbergm> heller: yep, thanks
<diehlpk_work> I advertised it on Twitter
hkaiser has quit [Quit: bye]
<heller> simbergm: I'd be happy to help out
<nikunj> aserio: yt?
<aserio> nikunj: yes
<nikunj> aserio: I've updated the result and benchmark section
<nikunj> and also updated the figure
<nikunj> could you write up the conclusion?
<nikunj> and also read through those sections once again?
<aserio> nikunj: I will definitely read through the extended abstract
<nikunj> aserio: thanks!
<aserio> nikunj: do you think I am the right person to write the conclusion?
<nikunj> we will still have to find places to cut short
<aserio> its kind of an important section
<nikunj> aserio: I can write up the conclusion myself, will you check for its correctness?
<aserio> I can do that
<nikunj> alright, I'll write it up rn. We will have to find places to delete
<nikunj> It's going to be 7 pages long otherwise
<aserio> Cutting is always easier than adding
<nikunj> that's true xD
<nikunj> alright I'll make the changes, please look into them when you have time
<nikunj> also let's do a call tomorrow at 10?
<nikunj> 10am your time
<aserio> I have a meeting scheduled at that time, but Hartmut may still be able to join
<aserio> I am editing Gabriel's paper now
<nikunj> ohh ok
<aserio> so I won't be able to get to your paper until tomorrow
<nikunj> I'll ask hartmut if he's free tomorrow
<nikunj> that's fine, do it when you have time
<aserio> diehlpk_work: see pm
hkaiser has joined #ste||ar
bita has joined #ste||ar
bita has quit [Client Quit]
<aserio> nikunj: btw, I am currently re-reading the extended abstract
<aserio> I am editing all sections but the Conclusion until you finish
<aserio> let me know if this will cause conflicts for you
<aserio> hkaiser: yt?
<aserio> K-ballo: you might be able to answer this one as well
<aserio> Is this correct to say: A dataflow waits for all provided futures to become ready,
<aserio> and then executes the specified function.
<aserio> Inside the dataflow, the implementation verifies the
<aserio> state of each future and then attaches the next future
<aserio> order that the futures become ready.
<aserio> as a continuation. This implementation may result in
<aserio> some variability of the cost of the dataflow depending on the
<hkaiser> aserio: sounds ok
<aserio> hkaiser: thanks!
<aserio> I have added it to nikunj 's paper
<hkaiser> nod, thanks
aserio has quit [Quit: aserio]
mreese3_ has quit [Ping timeout: 276 seconds]