When we (Neology) expanded our Windows Update Caching service, our chief propeller head (TheRodent) suggested a move from our old stalwart (apache2) towards the new kid on the block – NGINX.
Having had little prior exposure to the product I was rather hesitant but also keen so see its performance.
I just had a brief look at one of our busier accelerators and I must say NGINX really does ROCK:
| Date | Hits | Bandwidth |
|---|---|---|
| 09 Dec 2009 | 5 902 426 | 155.66 GB |
| 10 Dec 2009 | 3 664 433 | 97.11 GB |
| 11 Dec 2009 | 2 804 256 | 88.50 GB |
| 12 Dec 2009 | 1 986 264 | 70.81 GB |
Just under 6 million hits in a single day, not a bad thing 😉
Overall this is a bit of software that really does what it’s set out to be able to do – deal with high PPS and work reliably 😉
Only real downside so far is the limited features of the status module – would love to get more data from it so SNMP stats can become somewhat useful.