Solr loads an entire index into memory

I use solr for data like Name: Age: Gender: balance: nextbalance: percent

I have 30 M records on 4G on disk. I get by age: 23, which is only 50 entries. I indexed = "true" in the xml schema. Solr seems to load the entire index onto disk in memory (4G). Shouldn't he only retrieve 40 odd entries in memory?

+5
source share
3 answers

I think it depends on how you configure the cache (what it does and doesn't save in memory). Loading the entire index into memory can give you a huge performance boost in terms of the time it takes to get results, regardless of query.

:

+4

, . . solrconfig.xml?

<!-- documentCache caches Lucene Document objects (the stored fields for each document).
  -->
<documentCache
  class="solr.LRUCache"
  size="16384"
  initialSize="16384"/>
+3

, , , , . , 100% , . .

, , SOLR ( ) , . , , , . , SOLR , . solrconfig.xml.

Please note that this is a complex area and it will probably be difficult for you to fully understand caching if Google is your primary source of information. This is an area where it is best to study in one of the SOLR books.

+1
source

All Articles