alexandre-normand commented on PR #1654:
URL: https://github.com/apache/iceberg-go/pull/1654#issuecomment-5273491421

   I decided to scrap the plans for goccy/go-json given that we might be able 
to negate the poor performance impact of encoding/json from the [recently 
merged PR](https://github.com/apache/iceberg-go/pull/1749) from 
@technicolorbeat (we don't need to get the full snapshot list when loading 
tables and this is the bulk of the json that we have to decode). 
   
   I could have either completely closed this PR _or_ trimmed it down to only 
the benchmark that would be scoped _only_ to encoding/json benchmarking. I 
decided to do the latter since it was easy enough to do and you can make the 
call whether this can/should be merged or not worth it. 
   
   I could see value in keeping the benchmark around so that when 1.27 lands 
with encoding/json, we can review benchmark numbers before/after the golang 
upgrade and actually see how performance does improve. 
   
   Thinking ahead, I don't know if moving to using `encoding/json/v2` would be 
beneficial then vs just using `encoding/json` which will be backed with 
`encoding/json/v2` using v1 compatibility options. I haven't reviewed the 
details but it should be that just upgrading to 1.27 will be enough to bring 
performance benefits without requiring code changes. 
   


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to