The result of enabling thread pooling and then benchmarking the change shows some pretty impressive performance gains.
Now our server produces 9.5 Gbps, compared to ~1 Gbps without thread pools!
and
The average time to serve a 4-MB file has been reduced from 7.42 seconds to 226.32 milliseconds (33 times less), and the number of requests per second has increased by 31 times (250 vs 8)!
The title of the article seems wrong, as a best-case scenario has the possibility to increase bandwidth 10x and the time to serve a file is 33x faster (under the best of circumstances).
Worth a read and definitely worth a test in your lab: Thread Pools in NGINX Boost Performance 9x!