Pierre Salagnac created SOLR-18315:
--------------------------------------
Summary: Replicate CPU intensive UTF conversion tool by JVM calls
Key: SOLR-18315
URL: https://issues.apache.org/jira/browse/SOLR-18315
Project: Solr
Issue Type: Improvement
Components: SolrJ
Affects Versions: 10.0
Reporter: Pierre Salagnac
When sending updates, I noticed the code that converts Java strings (UTF16)
into UTF8 for {{JavaBinCodec}} is a CPU hotspot.
This may be either in ingesting clients or in Solr nodes when forwarding data.
All the code for UTF conversion in class {{ByteUtils}} is very CPU intensive.
The JVM provides methods for UTF conversion that are intrinsic candidates and
will be much more efficient, both in term of CPU and memory consumption.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
