aserio 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/
jaafar has joined #ste||ar
jaafar has quit [Ping timeout: 246 seconds]
ORion2 has joined #ste||ar
ORion2 has quit [Remote host closed the connection]
denis_blank has quit [Quit: denis_blank]
ORion2 has joined #ste||ar
EverYoung has quit [Ping timeout: 246 seconds]
Guest14201 has quit [Read error: Connection reset by peer]
hkaiser has joined #ste||ar
aserio has joined #ste||ar
ORion2 has quit [Remote host closed the connection]
bikineev has quit [Remote host closed the connection]
patg has joined #ste||ar
patg is now known as Guest99689
vamatya has quit [Ping timeout: 268 seconds]
eschnett has joined #ste||ar
K-ballo has quit [Quit: K-ballo]
aserio has quit [Quit: aserio]
vamatya has joined #ste||ar
Guest99689 is now known as patg
hkaiser has quit [Quit: bye]
ajaivgeorge_ has joined #ste||ar
pree has joined #ste||ar
<pree> patg : yt ?
<pree> I'm here :)
pree_ has joined #ste||ar
pree has quit [Read error: Connection reset by peer]
<heller_> pree_: consider different timezones
<pree_> heller_ : She said she will active around this time
<heller_> ok
<pree_> ok : )
<patg> pree_: I'm here
jaafar has joined #ste||ar
jaafar has quit [Ping timeout: 240 seconds]
vamatya has quit [Ping timeout: 240 seconds]
ajaivgeorge has joined #ste||ar
ajaivgeorge_ has quit [Read error: Connection reset by peer]
Matombo has joined #ste||ar
pree_ has quit [Ping timeout: 268 seconds]
<heller_> jbjnr: yt?
pree_ has joined #ste||ar
david_pfander has joined #ste||ar
<github> [hpx] sithhell created fixing_refcounting (+1 new commit): https://git.io/vHAxC
<github> hpx/fixing_refcounting 71ef3ee Thomas Heller: Properly releasing parcelport write handlers....
<github> [hpx] sithhell opened pull request #2698: Properly releasing parcelport write handlers. (master...fixing_refcounting) https://git.io/vHAxV
<github> [hpx] sithhell closed pull request #2692: Partially reverting changes to parcel_await (master...revert_parcel_await) https://git.io/vH9u9
pree_ has quit [Quit: AaBbCc]
pree_ has joined #ste||ar
heller_ has quit [Quit: http://quassel-irc.org - Chat comfortably. Anywhere.]
heller_ has joined #ste||ar
ajaivgeorge_ has joined #ste||ar
ajaivgeorge has quit [Ping timeout: 255 seconds]
ajaivgeorge_ has quit [Ping timeout: 255 seconds]
ajaivgeorge_ has joined #ste||ar
pree_ has quit [Ping timeout: 268 seconds]
pree_ has joined #ste||ar
Matombo has quit [Remote host closed the connection]
ajaivgeorge_ has quit [Ping timeout: 246 seconds]
ajaivgeorge has joined #ste||ar
ajaivgeorge is now known as ajaivgeorge_
mcopik has quit [Ping timeout: 260 seconds]
ajaivgeorge_ has quit [Ping timeout: 246 seconds]
ajaivgeorge has joined #ste||ar
ajaivgeorge is now known as ajaivgeorge_
ajaivgeorge_ has quit [Remote host closed the connection]
K-ballo has joined #ste||ar
hkaiser has joined #ste||ar
<github> [hpx] hkaiser deleted revert_parcel_await at b3e7c33: https://git.io/vHx3c
<heller_> hkaiser: hey, do you have a moment?
<hkaiser> heller_: sure
<hkaiser> g'morning!
<heller_> hkaiser: see pm
<heller_> good morning :D
bikineev has joined #ste||ar
hkaiser has quit [Read error: Connection reset by peer]
hkaiser_ has joined #ste||ar
hkaiser_ has quit [Read error: Connection reset by peer]
hkaiser has joined #ste||ar
Matombo has joined #ste||ar
elfring has joined #ste||ar
<github> [hpx] sithhell pushed 1 new commit to master: https://git.io/vHxny
<github> hpx/master 5a4e9e5 Thomas Heller: Merge pull request #2686 from STEllAR-GROUP/fixing_2682...
<github> [hpx] sithhell closed pull request #2698: Properly releasing parcelport write handlers. (master...fixing_refcounting) https://git.io/vHAxV
hkaiser has quit [Read error: Connection reset by peer]
bikineev has quit [Ping timeout: 246 seconds]
bikineev has joined #ste||ar
bikineev has quit [Ping timeout: 240 seconds]
<jbjnr> heller_: here for a moment or two
<heller_> jbjnr: see pm
<jbjnr> :?
<jbjnr> k
bikineev has joined #ste||ar
<heller_> jbjnr: ha! I got multiple succesful LF runs on 128 localities now
<heller_> nope, not really :/
hkaiser has joined #ste||ar
<hkaiser> ABresting: yt
<hkaiser> ?
<hkaiser> pree_: ?
<pree_> yes
<hkaiser> pree_: hey
<pree_> I'm here
<hkaiser> pree_: did parsa[w] get back to you now?
<pree_> Not yey
<pree_> *yet
<hkaiser> uhh
<hkaiser> I will have more time over the weekend and next week, I'll look into things for you
<hkaiser> please don't loose patience with us
<pree_> i got upset with domains
<hkaiser> upset? what do you mean?
<pree_> I can't able to relate with HPX. I'm totally lost for last 2 days
<hkaiser> pree_: that's a good sign ;)
<hkaiser> when you learn new things you first think you understand everything
<pree_> Are you making fun ??
<hkaiser> then you don't understand a thing anymore
<hkaiser> so this is progress
<pree_> haha that's really happening
<pree_> I didn't able to see any relationship between index-sets and components !
<hkaiser> pree_: nod, I can see that happening
<hkaiser> pree_: I'll be travelling today and will not be able to do anything
<hkaiser> but I'll get back to you tomorrow, promise
<pree_> I have some ideas to relate it with arrays. I don't know how to make it generic to support all containers .
<pree_> Okay hkaiser
<hkaiser> pree_: don't overthink and don't overdesign
<hkaiser> try to solve a simple very specialized problem first
<pree_> literally things goes out of scope when dealing with multi-dimensional arrays itself
<pree_> I have also look into
<pree_> ZPL regions
<hkaiser> what are zpl regions?
<pree_> ZPL is a language
<hkaiser> ahh
<pree_> regions specify the space
<hkaiser> yah, do some reading, that helps
<pree_> yes " space " it means physically
<heller_> pree_: components are just objects
<pree_> these are built-in's type (regions and space ) in a language level
<pree_> heller_ : where did you got this ideas ??
<pree_> and also how ?
<heller_> think of it that way: You want to create N objects. Those N objects need to be created somewhere
<heller_> which ideas?
<pree_> This project idea..
<heller_> from chapel :P
<heller_> as said, I want to say where those N objects should be located
<heller_> going with the matrix example...
<heller_> I have a Matrix, with MxM entries
<heller_> that is, I need to create N objects
<pree_> okay
<heller_> where N is M * M
<pree_> I'm here please
<pree_> proceed
<heller_> are you familar with linearising multi dimensional indices?
hkaiser has quit [Quit: bye]
<pree_> Not !
<heller_> bad
<heller_> so here is the problem
<heller_> I have N = M * M objects I want to allocate
<heller_> and I have L localities
<heller_> that is, I need to distribute my N objects onto L localities
<heller_> so I split up the N into different N_l where l goes from 0 to L-1
<heller_> makes sense
<heller_> ?
<pree_> no probelm i will learn it
<heller_> so I need a mapping going from the global indices to the local ones
<ABresting> hkaiser: yes
<heller_> try to understand this first
<pree_> Okay I will ...
<heller_> this is also nice
<pree_> If any probelm i will contact you
<pree_> This was the first pdf download for gsoc on april 4
<heller_> so this gives you a nice overview, right?
<heller_> at the very basis, is of course the linearization
<pree_> It gives but at the same time it also arises many questions
<heller_> did you ask those questions?
<pree_> By now i will go for your matrix example first
<pree_> No, I thought i will understand as time goes
<heller_> well, that's what mentors are for, aren't they?
<heller_> if you don't ask questions or discuss your problems, how should they know?
<heller_> another hint: take a step back, get some paper and pencil and do some drawings
<pree_> I will learn this linearisation and I will come back
<pree_> okay I will
bikineev has quit [Ping timeout: 246 seconds]
eschnett has quit [Quit: eschnett]
bikineev has joined #ste||ar
<jbjnr> heller_: 128 localities with LF : Did you change anything? I'm hoping you've tested the memory reduction thing
<heller_> memory reduction?
<jbjnr> max terminated threads nonsense
<heller_> the thread termination thingy?
<heller_> yeah, I have that, and reduced the number of preposts etc.
<heller_> makes things more smooth but not perfect yet...
<jbjnr> in what way not perfect
<jbjnr> ?
<heller_> still deadlocking
<heller_> I am suspecting a similar mess up as the one I just fixed
<jbjnr> hmmm. this is across all PPs. not good
<jbjnr> is that part of your LF branch that I still have not tested out?
<jbjnr> (sorry about that)
<jbjnr> I tihnk it was merged already too
jaafar has joined #ste||ar
ajaivgeorge has joined #ste||ar
ajaivgeorge_ has joined #ste||ar
ajaivgeorge_ has quit [Client Quit]
ajaivgeorge has quit [Ping timeout: 240 seconds]
eschnett has joined #ste||ar
patg has quit [Read error: Connection reset by peer]
EverYoung has joined #ste||ar
EverYoung has quit [Remote host closed the connection]
EverYoung has joined #ste||ar
david_pfander has quit [Ping timeout: 240 seconds]
bikineev has quit [Read error: No route to host]
bikineev has joined #ste||ar
bikineev has quit [Read error: Connection reset by peer]
bikineev has joined #ste||ar
pree__ has joined #ste||ar
pree_ has quit [Read error: Connection reset by peer]
Matombo has quit [Remote host closed the connection]
EverYoun_ has joined #ste||ar
EverYoun_ has quit [Remote host closed the connection]
EverYoun_ has joined #ste||ar
EverYoung has quit [Ping timeout: 246 seconds]
patg has joined #ste||ar
patg is now known as Guest30250
Guest30250 has quit [Read error: Connection reset by peer]
ajaivgeorge has joined #ste||ar
bibek_desktop has quit [Ping timeout: 258 seconds]
patg_ has joined #ste||ar
mcopik has joined #ste||ar
bibek_desktop has joined #ste||ar
jaafar has quit [Quit: Konversation terminated!]
EverYoun_ has quit [Ping timeout: 246 seconds]
aserio has joined #ste||ar
pree__ has quit [Ping timeout: 268 seconds]
aserio1 has joined #ste||ar
aserio has quit [Ping timeout: 240 seconds]
aserio1 is now known as aserio
hkaiser has joined #ste||ar
EverYoung has joined #ste||ar
EverYoung has quit [Remote host closed the connection]
EverYoung has joined #ste||ar
aserio has quit [Ping timeout: 268 seconds]
hkaiser has quit [Read error: Connection reset by peer]
bikineev has quit [Ping timeout: 246 seconds]
bikineev has joined #ste||ar
hkaiser has joined #ste||ar
akheir has joined #ste||ar
hkaiser has quit [Read error: Connection reset by peer]
hkaiser has joined #ste||ar
vamatya has joined #ste||ar
hkaiser has quit [Quit: bye]
<heller_> I'm taking bets on how many merges the current green survives
<jbjnr> where is this green of which you speak?
pree__ has joined #ste||ar
<zao> A state of the buildbot, wrought in mystery.
bikineev has quit [Ping timeout: 246 seconds]
bikineev has joined #ste||ar
<heller_> jbjnr: there is more and more coming
<jbjnr> I've seen more green in Donald Trump's environment plan than our buildbot
<zao> That's low.
<mcopik> ajaivgeorge: Ajai, how is the work going?
<heller_> jbjnr: getting better
<heller_> clean orange and stuff
jaafar has joined #ste||ar
<mcopik> there's definitely more orange than green in Herr Trump
bikineev has quit [Read error: Connection reset by peer]
bikineev has joined #ste||ar
thundergroudon[m has quit [Ping timeout: 240 seconds]
taeguk[m] has quit [Ping timeout: 255 seconds]
akheir has quit [Remote host closed the connection]
bikineev has quit [Remote host closed the connection]
<pree__> heller_ : I got something from chapel designer brad
<heller_> ok
<pree__> I hope it will help me a lot
<heller_> pree__: pretty sure it will
<heller_> so here is another misconception of yours
<pree__> What ??
<heller_> in Chapel, you have multidimensional arrays
<heller_> in C++, you only have one dimensional
<pree__> I looking for something like hpx::domains
<heller_> sure
<pree__> converting from domains to index would leave multidimension from arrays i guess so
<heller_> good idea, don't bother with the actual containers yet
<heller_> *nod*
<pree__> sorry I can't understand : )
<pree__> actual containers context
<heller_> try to implement domains first
<heller_> write testcases iterating over them etc
<heller_> then create the domain maps and generate the indices
<heller_> baby steps
<pree__> that's the plan , and using it arrays first
bikineev has joined #ste||ar
<jbjnr> what's the differnce between get_queue_length and get_thread_count for a tread queue
<jbjnr> what's the differnce between get_queue_length and get_thread_count for a tread queuethread queue
<jbjnr> grrrrr
<heller_> would have to look
<heller_> but IIRC, queue length is what is in the queue
<heller_> and the other is what had been executed
hkaiser has joined #ste||ar
<jbjnr> hkaiser: same question for you ...
<jbjnr> what's the differnce between get_queue_length and get_thread_count for a thread queue
<hkaiser> queue length is instantaneous, thread-count the overall number (cumulative
<hkaiser> no, hold on
<hkaiser> thread-count the number of active (non-terminated threads, instantaneous as well)
hkaiser has quit [Read error: Connection reset by peer]
hkaiser has joined #ste||ar
<jbjnr> by instantaneous you mean?
<jbjnr> at that moment running?
<hkaiser> jbjnr: yes
<hkaiser> just a snapshot of how many thread exists/are queued up
<pree__> heller_ : From- your statement " c++ have only one-dimensional arrray " ,Can I take jagged array as a example of that ?
<pree__> Or Do you specify anything else ?
<pree__> like skyline arrays
<zao> pree__: Apart from C-style [][][] arrays, you only have linear sequences for array-like storage.
<zao> std::vector<T>, std::array<T, N>, new T[N], etc.
<zao> If you have a multidimensional dataset of dimensions M,N, you can store that in a linear array of size M*N, by doing arithmetic on the indices.
<pree__> But zao but multidimensional one are actually stored linearly in memory
<zao> Doing it so that a particular (row,col) index corresponds to one linear index (i).
<zao> The point being is that you don't have any native multidimensional array type.
<hkaiser> pree__: it's not your task to think about multidimensional arrays
<hkaiser> distribution policies define the structure how partitions are distributed, not single elements
<pree__> understood zao thanks
<hkaiser> we can build arbitrary N-dimensional arrays on top of any 1d array using std::md_span or similar
<pree__> hkaiser : I tend to implement domains in hpx
<hkaiser> the point is to allow control over distributing _partitions_
<hkaiser> (for partitioned_vector)
<pree__> so i thought it will needed to deal with it
<hkaiser> if you think about creating a bunch of arbitrary components using distribution policies then there is no need to think about n-dimensionality either
<pree__> heller_ : Advised to implement domains first and then domain maps
<hkaiser> ok
<hkaiser> sounds good to me, however - again - I'd rather see some human-readable description of what you want to implement before you even start coding
<pree__> haha okay !
<pree__> But tomorrow it's too late now
hkaiser has quit [Quit: bye]
patg_ has quit [Quit: This computer has gone to sleep]
bikineev has quit [Remote host closed the connection]
eschnett has quit [Quit: eschnett]
thundergroudon[m has joined #ste||ar
elfring has quit [Quit: Konversation terminated!]
taeguk[m] has joined #ste||ar
pree__ has quit [Ping timeout: 268 seconds]
pree__ has joined #ste||ar
pree__ has quit [Client Quit]
diehlpk_work has quit [Quit: Leaving]
EverYoung has quit [Remote host closed the connection]
EverYoung has joined #ste||ar
bikineev has joined #ste||ar
EverYoung has quit [Ping timeout: 255 seconds]
bikineev_ has joined #ste||ar
bikineev has quit [*.net *.split]
mcopik has quit [*.net *.split]
mcopik has joined #ste||ar
EverYoung has joined #ste||ar