I've read a lot of ES Performance Guides and most of them suggest setting
the ES_HEAP_SIZE to half the RAM, but not more than 32 GB.
ES is running on a Server with 32 GB of RAM, with nginx, php5-fpm and mysql
on this machine and i'm only indexing around 10.000 Documents with a total
size of 5 MB.
The Index will probably grow to around 20.000 - 30.000 (10 - 15 MB)
Documents in the near future, but thats - compared to other clusters - is
still a fairly small amount of data.
I dont have heavy traffic nor i'm doing crazy searches.
Most types are not_analyzed, because i'm primarily using ES for filtering,
aggregations and sorting, full text search is just a minor benefit for me,
but it would be crazy doing all the filtering with mysql and not with ES.
Setting the ES_HEAP_SIZE to around 16 GB seems a bit of an overkill for my
usecase, but since i'm using filters (which are cacheable) and doing
aggregations with sorting, setting the ES_HEAP_SIZE at around 2 GB feels
kinda right for me, but i'm not sure if i'm missing something here.
Any reccommendation on the ES_HEAP_SIZE with a small number of documents is
really appreciated, thanks!
I've read a lot of ES Performance Guides and most of them suggest setting the ES_HEAP_SIZE to half the RAM, but not more than 32 GB.
ES is running on a Server with 32 GB of RAM, with nginx, php5-fpm and mysql on this machine and i'm only indexing around 10.000 Documents with a total size of 5 MB.
The Index will probably grow to around 20.000 - 30.000 (10 - 15 MB) Documents in the near future, but thats - compared to other clusters - is still a fairly small amount of data.
I dont have heavy traffic nor i'm doing crazy searches.
Most types are not_analyzed, because i'm primarily using ES for filtering, aggregations and sorting, full text search is just a minor benefit for me, but it would be crazy doing all the filtering with mysql and not with ES.
Setting the ES_HEAP_SIZE to around 16 GB seems a bit of an overkill for my usecase, but since i'm using filters (which are cacheable) and doing aggregations with sorting, setting the ES_HEAP_SIZE at around 2 GB feels kinda right for me, but i'm not sure if i'm missing something here.
Any reccommendation on the ES_HEAP_SIZE with a small number of documents is really appreciated, thanks!
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.