High DISK (SSD) Usage While Optimizing - page 2
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Please show your Strategy Tester settings, a screenshot will do the job.
Is your EA single symbol or multi-symbols ?
Screenshots provided below.
It's a single symbol EA.
I added some important info in the comment #8 and the screenshot of the error in the log.
Screenshots provided below.
It's a single symbol EA.
I added some important info in the comment #8 and the screenshot of the error in the log.
So it's 10 years of custom ticks. How much ticks is that ?
Anyway it's a lot of needed memory. How much free memory do you have before starting the optimization ?
Assuming you are testing on all 12 threads of the 6 cores, with the full 32GB RAM, that is a maximum of 2.67GB available for each thread.
That is not enough RAM for 10 years of highly packed tick data on a Custom Symbol, especially if your EA is inefficient. So, ...
... reduce the number of threads being used for optimisation and reduce the test period, to a point where the RAM usage does not go beyond 80% maximum usage.
You need some leeway for the system and for the disk cacheing to work in your favor and not against you.
NEW IMPORTANT INFORMATION
I bought a better SSD (Samsung 990 Pro 2TB) - The difference is that now the CPU doesn't go to 0% when the SSD is working, but the optimization is still not progressing.
Return that SSD, it's not needed at all, you need more RAM, not faster SSD.
Assuming you are testing on all 12 threads of the 6 cores, with the full 32GB RAM, that is a maximum of 2.67GB available for each thread.
That is not enough RAM for 10 years of highly packed tick data on a Custom Symbol, especially if your EA is inefficient. So, ...
... reduce the number of threads being used for optimisation and reduce the test period, to a point where the RAM usage does not go beyond 80% maximum usage.
You need some leeway for the system and for the disk cacheing to work in your favor and not against you.
I tested on 12, 6 and 1 agent. Tried less than 10 years and the problem is still there. If I go for less than 5 years it works, but it's not the goal.
My EA is not inefficient, and the 3 other developers EA's aren't too.
With 30% RAM usage the problem is still there, it seems like the amount of years are the problem.
THIS DIDN'T HAPPEN BEFORE - It all started frist days of December. So It's not a 100% hardware issue. Im sure theres more to it. I've seen some other people post about it.
I tested on 12, 6 and 1 agent. Tried less than 10 years and the problem is still there. If I go for less than 5 years it works, but it's not the goal.
My EA is not inefficient, and the 3 other developers EA's aren't too.
With 30% RAM usage the problem is still there, it seems like the amount of years are the problem.
THIS DIDN'T HAPPEN BEFORE - It all started frist days of December. So It's not a 100% hardware issue. Im sure theres more to it. I've seen some other people post about it.
Once upon a time, i did backtest of same ea i use today, it only used 20gig, but then 30, and now 43. do not assume that you do not need more ram just because it worked a few days or weeks ago. There as been a few updates of mtx in that time.
I tested on 12, 6 and 1 agent. Tried less than 10 years and the problem is still there. If I go for less than 5 years it works, but it's not the goal.
My EA is not inefficient, and the 3 other developers EA's aren't too.
With 30% RAM usage the problem is still there, it seems like the amount of years are the problem.
THIS DIDN'T HAPPEN BEFORE - It all started frist days of December. So It's not a 100% hardware issue. Im sure theres more to it. I've seen some other people post about it.
If you don't answer the questions you have been asked, nobody can help.
Just run a single backtest (not an optimization) and send me the log...if you want.
So it's 10 years of custom ticks. How much ticks is that ?
Anyway it's a lot of needed memory. How much free memory do you have before starting the optimization ?
Don't know how many ticks, a lot probably.
1.8 TB Before starting optimization.
The single backtest log is attached below.
Thanks for helping.
Other stuff that happens now and before didn't': When visual backtesting, the first trade that is opened freezes the tester for 15 seconds, and then it continues with the ticks and moving till the next tradeDon't know how many ticks, a lot probably.
1.8 TB Before starting optimization.
The single backtest log is attached below.
Thanks for helping.
Other stuff that happens now and before didn't': When visual backtesting, the first trade that is opened freezes the tester for 15 seconds, and then it continues with the ticks and moving till the next tradeYou forgot to say you are using a Mac ! Not sure what is the impact, I don't have one to check.
QI 0 00:59:32.355 Core 01 USDJPY Dukascopy FTMO,M5: 116358536 ticks, 355502 bars generated. Environment synchronized in 0:00:00.132. Test passed in 0:00:39.958 (including ticks preprocessing 0:00:09.905).
EN 0 00:59:32.355 Core 01 USDJPY Dukascopy FTMO,M5: total time from login to stop testing 0:00:40.090 (including 0:00:03.489 for history data synchronization)
QQ 0 00:59:32.355 Core 01 382142260 total ticks for all symbols
QE 0 00:59:32.355 Core 01 USDJPY: generate 265893602 ticks in 0:00:05.515, passed to tester 265783727 ticks
IJ 0 00:59:32.355 Core 01 USDJPY Dukascopy FTMO: generate 116358536 ticks in 0:00:04.390, passed to tester 116358537 ticks
Not far of 400 millions ticks...
MG 0 00:59:32.355 Core 01 8280 Mb memory used including 255 Mb of history data, 7360 Mb of tick data
And around 16 GB for the backtest only. It's all normal.
P.S: Not related to the RAM/Disk issue, but your trading log is full of errors. Also the trades stopped in 2019 for a backtest supposed to run up to December 2024.