Express: any plan on performance improvements?

Created on 6 May 2017  路  5Comments  路  Source: expressjs/express

I have tested koa and express with basic 'Hello world' response.

performance comparison

where koa is much faster:

koa's records:

Requests per second:    4345.20 [#/sec] (mean)
Requests per second:    4545.75 [#/sec] (mean)
Requests per second:    4829.70 [#/sec] (mean)
Requests per second:    4515.21 [#/sec] (mean)

express's records:

Requests per second:    3572.34 [#/sec] (mean)
Requests per second:    4021.62 [#/sec] (mean)
Requests per second:    3913.94 [#/sec] (mean)
Requests per second:    3842.52 [#/sec] (mean)

vig

vig is a new framework focusing on web logic which is usually called controller by most mvc-based frameworks.

It is designed to be performance aware, modular, simple and re-entrant.

and as i have tested, there is no big difference in performance with express.

vig's records:

Requests per second:    3563.68 [#/sec] (mean)
Requests per second:    3882.52 [#/sec] (mean)
Requests per second:    3783.89 [#/sec] (mean)
Requests per second:    3921.03 [#/sec] (mean)

since express is much slower now, I am wondering if there is any plan on performance improvements?

(for koa is still under development and somewhat off the road to me, i may not set time to add support to koa for the time being.)

question

Most helpful comment

I have already implemented all the improvements I know about.

All 5 comments

We are always working on improving performance. If you find anything specific to change / improve, please feel free to make a pull request or outline where the bottleneck is and a suggested solution :)

I am afraid I haven't got time at time being.
I would be glad to hear improvements from you:)

I would plan to add an adapter for koa if problem remains, thought it seems ugly to me :-)

I have already implemented all the improvements I know about.

sad to hear that.

We are always working on improving performance. If you find anything specific to change / improve, please feel free to make a pull request or outline where the bottleneck is and a suggested solution :)

Was this page helpful?
0 / 5 - 0 ratings