Swoole Performance Benchmark ...

[Fakhar Anwar]

Hi,

For swoole;

Query-1: Did you use http server of swoole to test performance?

Query-2: Did you use Swoole built-in coroutines like those for MySQL?

Query-3: Did you use Swoole with its http-2 configuration ?

Query-4: What is wrong with Swoole Performance as it is showing degraddation over years, here is my observation :

Some older benchmark (Date not visible)

Ranked on number 4, with ...

304,264 rps (stays above aspcore 299,545)

https://www.techempower.com/benchmarks/#section=test&runid=9d5522a6-2917-467a-9d7a-8c0f6a8ed790

---

2017

Ranked on number 3, with ...
303,514 rps (performance stays above aspcore 294,285)

https://www.techempower.com/benchmarks/#section=data-r17

2018

Ranked on number 11,  with ....
307,673 (performance stays above aspcore 300,613)

https://www.techempower.com/benchmarks/#section=data-r18

2019

Ranked on number 25, with ...
283,728 rps (performance far below php-ngx-pgsql 356, 507 rps, and one step below asp.net core 285,398 rps)

https://www.techempower.com/benchmarks/#section=data-r19&hw=ph&test=fortune

What i know about Swoole is that it loads all files in RAM only once plus it does not require a separate Web Server and Fast-CGI Server to work with.

Looking forward to hearing from you in this regard,

Thanks in anticipation.

Kind Regards,

Fakhar Anwar