Greetings,
I am sorry for having another problem here. I ran a fairly small scale experiment (80 participants at the same time) with few background calculations on Heroku but the pages were loading extremely slow. I used Standard 2X web dynos, Standard 2 Postgres and Hobby Redis, which seems to be enough for this purpose.
The metric seems to show that both the response time and dyno load are very high (figures below shows the detailed statistics). Also, the logfile shows that there is something wrong with the postgres. I am not sure whether this is due to the limitation of the dyno or postgres I am using now or there are bugs in my code.
Could you please provide some advice on which kind of bug in the code would cause this kind of problem? Thank you so much in advance for your help!!
Best wishes,
Jindi
