You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Right now the API on works when running with a single worker thread. This could cause slowdowns when there are many requests at once.
NOTE: When content is actually downloaded, a task is added to a thread pool to perform those actions. This thread pool is configured as None which will use the number of logical CPU's available. This means that only API requests could cause a bottleneck.
Describe the proposed solution
Update the API codebase to support multiple worker threads. This could be done by using some sort of shared memory, Redis cache, or database. Implementation does not really matter, assuming that it works and is not causing another bottleneck that is worse than current functionality.
Alternatives considered
N/A
Importance
nice to have
Additional information
This will require that the current session manager class is available for all worker thread. This is to ensure that one users session information is available if many worker threads are process different requests by that user. Without this we could see duplicate content requests, failed requests, etc.
The text was updated successfully, but these errors were encountered:
Describe the problem
Right now the API on works when running with a single worker thread. This could cause slowdowns when there are many requests at once.
Describe the proposed solution
Update the API codebase to support multiple worker threads. This could be done by using some sort of shared memory, Redis cache, or database. Implementation does not really matter, assuming that it works and is not causing another bottleneck that is worse than current functionality.
Alternatives considered
N/A
Importance
nice to have
Additional information
This will require that the current session manager class is available for all worker thread. This is to ensure that one users session information is available if many worker threads are process different requests by that user. Without this we could see duplicate content requests, failed requests, etc.
The text was updated successfully, but these errors were encountered: