Hi ,
I have an Inventory application created in Quarkus framework with AWS RDS Db. I configured H2 inmemory db with the application for faster performance. on my local inventory table in RDS Db was having 1 lc records so h2 inmemory while startup of application was building this table in memory successfully. I did load tests , http tests with wrk tool, JMH tools with success.But when I took this to production server where I am having 6lc records in inventory on RDS. H2 inmemory is not able to load inventory, build H2 in mem .
If I use MVStore=false in my connection url for in memory h2 db, I am able to build the db in mem successfully, but soon starts giving heapspace and thread blocking issues. Also if I run it on File server mode it loads quickly but I want to run it in memory.
This is my connection url : quarkus.datasource.jdbc.url=jdbc:h2:mem:INVDB;MODE=MYSQL;IGNORECASE=TRUE;CACHE_SIZE=6368709;MV_STORE=false;RETENTION_TIME=1000
please suggest how to load in memory this database successfully with no heap space issues.
With regards,
Tapan Aggarwal