I have several microservices written in .net core running on Kubernetes. The only way I can get those services to be performant is to set a minimum thread count of 300. Else, those services end up with tons of Redis timeout errors, problem connecting to SQL Server and resource unavailable errors connecting to external services. Considering that Microsoft recommends to allow the platform to manage the minimum thread count, what are the possible ramifications of setting the minimum thread to such a high value? I ran several substantial load tests including one with 10,000 concurrent users and all the services performed excellently with the minimum thread count set manually, without errors. P.S. I have both gRPC and Web API services.

Although everything looks good, I want to make sure that I am not setting myself up for trouble in production.