GPU support

GPU support should be added for performance.
Attachments
  • Votes +27
  • Project StrategyQuant X
  • Type Feature
  • Status Archived
  • Priority Normal
  • Assignee None
  • Milestone Archived (To be done later)

History

OS
#1

trader4711

10.10.2018 21:11
Voted for this task.
KO
#2

KrzysztofOzog

11.10.2018 07:24
Voted for this task.
FM
#3

Aimak_Rokalno

13.10.2018 12:42
OpenCL?
t
#4

tommmmmm

14.10.2018 16:11
As far as I remember Mark said this will be added one day, but not anytime soon.
TT
#5

Tamas

15.10.2018 15:00

Milestone changed from Build 112 to Build 113

a
#6

Ash24FX

16.10.2018 20:02
Voted for this task.
MF
#7

Mark Fric

26.10.2018 11:26

Milestone changed from Build 113 to Build 114

b
#8

bentra

30.10.2018 00:09
Voted for this task.
MF
#9

Mark Fric

16.11.2018 15:23

Milestone changed from Build 114 to To be done later

KL
#10

kainc301

03.12.2018 15:54

This may be a helpful reference for devs regarding this topic


https://www.youtube.com/watch?v=CIjdipU66qw


Rr
#11

Partizanas

03.12.2018 19:27
Voted for this task.
rg
#12

Robert Gergelyi

04.12.2018 16:46
Voted for this task.
o
#13

Enric

06.12.2018 10:21
Voted for this task.
l
#14

Laur2000

16.12.2018 23:17
Voted for this task.
DV
#15

TheLibertarianTrader

21.12.2018 23:03
Voted for this task.
jn
#16

Josef Němeček

28.12.2018 14:46
Voted for this task.
KL
#17

kainc301

05.01.2019 01:08
Voted for this task.
f
#18

felipebr

19.05.2019 15:31
Voted for this task.
SB
#19

sbecm

23.05.2019 23:31
Voted for this task.
KL
#20

kainc301

18.08.2019 23:32
I recently stumbled upon two academic researchers who used CUDA libraries in order to integrate GPU processing with genetic training for algorithms trained on FX markets. I highly recommend taking a look at their methodology as this was previously considered impossible to integrate into genetic algorithms http://people.scs.carleton.ca/~dmckenne/5704/Paper/Final_Paper.pdf
KL
#21

kainc301

19.08.2019 16:51
Also wanted to note that just recently, CUDA was used in order to achieve 6000x performance increase on NVIDIA GPUs backtesting financial markets. I don't know if the libraries for CUDA were updated to better handle this type of computation but it has been achieved. There should be a way to introduce GPU computing across the board for different processes in SQX. 


https://www.nvidia.com/content/dam/en-zz/Solutions/industries/finance/finance-trading-executive-briefing-hr-web.pdf

t
#22

tnickel

20.08.2019 02:39
Voted for this task.
t
#23

tnickel

20.08.2019 02:39
This will be nice, but very difficult 
MF
#24

Mark Fric

16.02.2020 13:52

Status changed from New to Archived

AG
#25

Tukirito

16.02.2020 20:34
Voted for this task.
WH
#26

Stormin_Norman

08.03.2020 15:25
Voted for this task.
m
#27

Mariano

30.03.2020 16:55
Voted for this task.
SF
#28

SergeyF

10.08.2020 01:52
Voted for this task.
k
#29

Karish

16.10.2020 22:30
Voted for this task.
L
#30

Lumen2730

08.11.2020 19:54
Voted for this task.
f
#31

FirestarZA

24.03.2022 18:25
Voted for this task.
w
#32

wiraputu

30.11.2022 14:19
Voted for this task.
t
#33

tofu

27.03.2023 09:57
Voted for this task.
E
#34

Emmanuel

16.09.2023 13:45
Voted for this task.
CY
#35

itgkwyc@gmail.com

03.12.2023 09:37
Voted for this task.
CY
#36

itgkwyc@gmail.com

03.12.2023 09:47

Attachment 1701592950947.jpeg added

1701592950947.jpeg
(64.76 KiB)
"I use Python + CUDA to create a backtesting system, which can output backtesting results for 38,000 different indicator parameters in about 150 seconds. This is much more efficient compared to SQ's CPU utilization."

Votes: +27

Drop files to upload

or

choose files

Max size: 5MB

Not allowed: exe, msi, application, reg, php, js, htaccess, htpasswd, gitignore

...
Wait please