Is TypeScript (NodeJS) Faster than Go?? | A server comparison

…call it an MVP publicly because why would you. I know because I've worked at successful companies with MRR in the millions that did this. You are aware of the tech debt and at some point you fix it. > Check these legendary videos which literally compares GO vs Node.js in place where Node.js is strong at i.e. IO and see how far it falls behind GO link to this video and link to this video Like I said, in raw performance Go is gonna beat Node/JS any day, that's not the case I'm making, those videos are not comparing real situations with real access patterns. >no one I know is trying to build servers in rust instead of GO eventhough you can try to do that Well and I do know people doing just that, so agree to d…

…k is somewhere else But the bottleneck was in Node.js is what was the point and proven in flamegraph and usecase. That's why a replacement with GO gave then 3x more performance with MUCH less memory. Check these legendary videos which literally compares GO vs Node.js in place where Node.js is strong at i.e. IO and see how far it falls behind GO [video and [video >Now in the last couple of years, that trend seems to be making a comeback, only this time it's about moving from Go to Rust. And I wonder the same, are they really hitting the limits of the Go runtime or is there some architectural issue they missed. Although this point is also answered in the above mentioned videos but stil…

Well, Go HTTP/WS servers in general are faster than Node (I think we both agree on that, right?). The question is "how much faster?" Beside benchmarks here is a good video about the topic (he measured WS servers). As you can see the "ws" package throughput didn't stand a chance against Go on bigger connectivity (1500 clients to a single WS server as shown in the video). You can argue that "I can always put 10 more WS servers so..." but it will cost you (or your company) money and I am trying to *reduce* the cost not to increase it.