Compare the performance of Java serialization vs Jackson CBOR vs Borer vs Kryo

[PawelLipski]

Each of Jackson/Borer/Kryo claims to be faster than Java serialization... let's do a simple check (a few million calls of SerializationTestKit#verify?) how do they actually compare. Let's consider a few different class objects, with ADTs/Seqs/Maps involved.

[PawelLipski]

(rel: sirthias/borer#37)

[MarconZet]

Maybe @HubertBalcerzak should do it? Regardless of what I do, this task will remain relevant.

[PawelLipski]

Makes sense, let's discuss on a call today...

[PawelLipski]

To check: https://github.com/maxbundscherer/akka-serialization-comparison

[plokhotnyuk]

Here is a comprehensive comparison of different JSON parsers for Scala on different JVMs using different data types and real-world samples (sources).

Also the following PR has sources and results of comparison some JSON parsers for Scala with binary parsers for Scala /Java including Java serialization: dkomanov/scala-serialization#8

[PawelLipski]

Thanks @plokhotnyuk ! let us take a look...

[MarconZet]

@plokhotnyuk is it possible to create a link to plokhotnyuk.github.io/jsoniter-scala/ in a way, to only show specific subset of data?

[plokhotnyuk]

@MarconZet As an option you can save it locally and remove unwonted data from provided.js.

A better option would be just running only required benchmarks on your env. and then using JMH Visualizer site to plot charts or compare results.