A boss of mine is not convinced to use Elixir for our production apps. Another round of our polemics was the test. He wrote two small apps using Vert.x and Elixir. They render static text and real data from MongoDB serialized into JSON. For static response Vert.x performed about 10% better. Also…

Having a quick glance, I can see two things: you only have 10 database connections opened (that’s the default). Depending on the concurrency of the test, it might be a good idea to open some more. Another place you could optimise is to try a NIF-based JSON encoder (like jiffy) instead of Poison.

[image] michalmuskala: Another place you could optimise is to try a NIF-based JSON encoder (like jiffy) instead of Poison. Especially this since MongoDB is so oddly designed in that it uses so much JSON… Really should not use MongoDB…

The optimization discussion is always interesting, but… does a 10 or 20% difference in speed really matter here? You’re basically choosing between the JVM vs BEAM. I think the difference is close enough that you should consider maintainability, learning curve, debuggability, whether you’ll need clus…

Thank you [image] michalmuskala: Another place you could optimise is to try a NIF-based JSON encoder (like jiffy) instead of Poison. On wrk -t20 -c100 -d15s it improved from 13k to 16k requests.

I wrote some tips in this post . It’s about Phoenix, but some points can be applied to plain cowboy: Bench an OTP release built for production. Change the setting of max_keepalive option. The default of 100 means that cowboy is dropping the connection after it serves 100 requests, so there will be …

Thank you for so broad reply. [image] sasajuric: Then you can try the same thing with Vert.x and see the behaviour. Assuming you properly configure just one worker OS thread, I’d be very surprised if Vert.x wasn’t completely blocked by the infinite request. I did not tried this yet, but the …

just after a quick look: (might contain mistakes/misunderstandings on my part) elixir is benchmarked through a router while vertx is not. the db query for elixir has limit 20 while the vertx has limit 10 (twice the data/serialization?). the db pools are not identical. vertx has the default 100 li…

[image] mpugach: I did not tried this yet, but the case is mentioned in their documentation. To be fair there is a workaround. Oh I’m positive there’s always a workaround, even if it’s not explicitly supported by the library itself. Worse comes to worst, you can always start such activity in …

Vert.x is a polygot implementation of the node paradigm on the JVM. You can spawn ‘verticles’ to handle a single API endpoint which all communicate through a messaging backend that extends to the clients Think node + messageMQ perhaps. They do have some good abstractions to reduce callback hell. I…

Elixir synthetic performance test

Questions & Help Questions

mpugach February 7, 2017, 5:23am 5

Thank you

On wrk -t20 -c100 -d15s it improved from 13k to 16k requests.