Re:Sinatra::Synchrony - Evented, fast, concurrent Ruby web a...
by shapeshed on Sunday December 18, 2011 @05:52PM
(#3367764)
performance benchmark using the same ab tests for node http://tjholowaychuk.com/post/543953703/express-vs-sinatra-b...
[ Parent
| on HN ]
Re:Sinatra::Synchrony - Evented, fast, concurrent Ruby web a...
by rgbrgb on Sunday December 18, 2011 @06:00PM
(#3367781)
So, if I'm reading this right, response time and reqs/second look pretty comparable between this and express.
Re:Sinatra::Synchrony - Evented, fast, concurrent Ruby web a...
by stephth on Sunday December 18, 2011 @07:43PM
(#3367985)
I think you're reading it right on the plain Hello World benches, but the in the HAML and SASS benches node is already 2 to 4 times faster with no concurrency. I think this speaks a lot for the impressive speed of V8. HAML/SASS are originally Ruby libraries, and carry a few years of experience on their backs, unlike their javascript versions. These V8 numbers are even (far) more impressive: http://attractivechaos.github.com/plb
Disclaimer: I use Ruby and have no experience with Node.js.
(edit: improved text)
Re:Sinatra::Synchrony - Evented, fast, concurrent Ruby web a...
by ScottMFisher on Sunday December 18, 2011 @06:54PM
(#3367881)
That's just plain old Sinatra.
Re:Sinatra::Synchrony - Evented, fast, concurrent Ruby web a...
by rgbrgb on Sunday December 18, 2011 @07:14PM
(#3367914)
True but you can compare the Express benchmarks to those mentioned on the Synchrony site.
Re:Sinatra::Synchrony - Evented, fast, concurrent Ruby web a...
by mrinterweb on Sunday December 18, 2011 @08:33PM
(#3368086)
Those benchmarks are not leveraging the concurrency that Sinatra::Synchrony brings to Sinatra. All this benchmark shows is that a concurrent application can serve out more requests than a non-concurrent application when tested concurrently.