gus-asf commented on PR #2447:
URL: https://github.com/apache/solr/pull/2447#issuecomment-2104536235

   > > I'm not so keen on the loss of 
https://issues.apache.org/jira/browse/SOLR-8349 along with this. It was only 
tied to the blob repository because that seemed to be how folks wanted me to 
load resources at the time.
   > 
   > Can you share more about how this is being used? I took a quick looksee at 
[SOLR-8349](https://issues.apache.org/jira/browse/SOLR-8349) and it points at 
[SOLR-3443](https://issues.apache.org/jira/browse/SOLR-3443) which appears to 
never had been completed.
   > 
   > I do think thinking about sharing memory might be something we need to 
have a good handle on for our future ML powered world.. As we add more models 
to our query/ingest steps, then maybe we would want to share memory? (Dunno how 
that would all work!).
   > 
   > If we move forward with this commit, we can always restore the code from 
Github right? And use our FilestoreAPI or other newer APIs?
   
   Yeah I just noticed that I never circled back to 3443, but when I made the 
contribution the original patch (not using blobs) was used by a client because 
they had 40 core server that wanted to load a 1 GB post-code to lat/long 
dictionary (US & UK) in a custom component that was tagging documents with 
lat/long based on post codes. Loading it once rather than 40 times was pretty 
important. At a later date I think they converted this to the blob version to 
upgrade, but I was elsewhere by that time.


-- 
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