Are there any special tools for identifying discourse performance bottlenecks?
Our forecast is that next year the number of our users will triple, so from today, we must provide the necessary resources for this scale.
