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
K-ballo has quit [Quit: K-ballo]
zao_ has joined #ste||ar
zao has quit [Ping timeout: 268 seconds]
zao_ is now known as zao
hkaiser has quit [Quit: Bye!]
zao has quit [Ping timeout: 260 seconds]
zao has joined #ste||ar
tufei_ has quit [Remote host closed the connection]
tufei_ has joined #ste||ar
pedro_barbosa[m] has quit [*.net *.split]
pedro_barbosa[m] has joined #ste||ar
srinivasyadav227 has quit [*.net *.split]
KordeJong[m] has quit [*.net *.split]
LorenDB[m] has quit [*.net *.split]
KordeJong[m] has joined #ste||ar
LorenDB[m] has joined #ste||ar
srinivasyadav227 has joined #ste||ar
FunMiles has joined #ste||ar
FunMiles has quit [Ping timeout: 250 seconds]
tufei_ has quit [Remote host closed the connection]
tufei_ has joined #ste||ar
K-ballo has joined #ste||ar
hkaiser has joined #ste||ar
<jedi18[m]>
hkaiser: Even with the adding random elements to a global sum, I'm still getting similar speedups in the benchmarks of a factor of 500 for replace_if and 300 for reverse :/
<jedi18[m]>
Btw shall I try running these benchmarks for the stl parallel algorithms as well to see if we're getting a similar speedup there?
<hkaiser>
that doesn't sound right
<hkaiser>
jedi18[m]: sure, try doing that - it's a good excercise ...
<jedi18[m]>
I'll try using google benchmark or something this time, just for a variation in the benchmarking method
<hkaiser>
also, you can try making the global 'volatile'
<hkaiser>
yes, those libraries usually have their own precautions against optimization
<jedi18[m]>
Oh ok thanks, I'll also try more complicated uses of the result vector
<jedi18[m]>
Also, regarding the remove_if parallel algorithm, looks like we are using a sequential third stage so that the algorithm is in place https://github.com/STEllAR-GROUP/hpx/issues/2804
<jedi18[m]>
I'll try adding a different one using extra memory using the new scan partitioner and see how it performs in comparision
<hkaiser>
ok, makes sense
<hkaiser>
jedi18[m]: btw, I just closed #1141 and #1668!
<hkaiser>
thanks again for all the help you provided!
<jedi18[m]>
Yay!! :D, glad to help!
<jedi18[m]>
I guess min,max,minmax is the one holding up the remaining issues?
<hkaiser>
yes
<hkaiser>
it's the next thing on my list
<jedi18[m]>
Thanks!, I'll try looking into too though I haven't been able to make much progress there
hkaiser has quit [Quit: Bye!]
FunMiles has joined #ste||ar
diehlpk_work has joined #ste||ar
<K-ballo>
vs2022 is out
hkaiser has joined #ste||ar
FunMiles has quit [Remote host closed the connection]
K-ballo has quit [Quit: K-ballo]
K-ballo has joined #ste||ar
FunMiles has joined #ste||ar
hkaiser has quit [Ping timeout: 268 seconds]
FunMiles has quit [Ping timeout: 250 seconds]
FunMiles has joined #ste||ar
FunMiles has quit [Ping timeout: 240 seconds]
FunMiles has joined #ste||ar
FunMiles has quit [Remote host closed the connection]
hkaiser has joined #ste||ar
FunMiles has joined #ste||ar
FunMiles has quit [Ping timeout: 240 seconds]
FunMiles has joined #ste||ar
hkaiser has quit [Quit: Bye!]
FunMiles has quit [Ping timeout: 240 seconds]
Coldblackice_ has joined #ste||ar
Coldblackice has quit [Ping timeout: 260 seconds]
FunMiles has joined #ste||ar
FunMiles has quit [Ping timeout: 244 seconds]
FunMiles has joined #ste||ar
FunMiles has quit [Ping timeout: 244 seconds]
hkaiser has joined #ste||ar
FunMiles has joined #ste||ar
FunMiles has quit [Ping timeout: 250 seconds]
Coldblackice has joined #ste||ar
FunMiles has joined #ste||ar
Coldblackice_ has quit [Ping timeout: 240 seconds]
FunMiles has quit [Remote host closed the connection]