4 ms·
Comparing comet performance between node and ringo should be interesting. Ringo uses jetty/cometd, so it should scale pretty decently as well. My guess is that
by hannesw 16y ago
Comparing comet performance between node and ringo should be interesting. Ringo uses jetty/cometd, so it should scale pretty decently as well. My guess is that node will use less memory per connection (and thus allow more simultanous connections), while ringo/jetty should have better latency. This is also what this benchmark suggests: http://praxx.is/post/486034949/comet-with-bayeux-node-js-vs-jetty-and-cometd http://praxx.is/post/486034949/comet-with-bayeux-node-js-vs-...
- amix 16y agoBefore our node.js solution we actually ran Java along with a highly optimized JBoss Netty comet implementation and it had a hard time scaling since it used _a lot_ of memory. Generally, our Java solution used around 10 times more memory than our node.js solution. Like I note in that post: "We have rewritten to node.js from Java+JBoss Netty, because the old solution had scalability problems. We have also been able to build a very sophisticated messaging system given JavaScript expressiveness and the simplicity of node.js. Generally, we have found node.js and V8 very impressive - and it uses about 10 times less memory than Java+Netty did." Jetty Comet uses a lot more memory than a vanilla JBoss Netty implementation - at least based on my benchmarks.
- hannesw 16y agoI know, i read that post. I'm planning to do a Node/Ringo comet comparison soon, and memory usage/max connections will be a part of that.
- praxxis 16y agoPlease do. I don't know why people keep reposting my benchmark, its terrible and really only applied to what I'm doing with Comet. Memory, etc, isn't the biggest factor for us.