Hi Ray, Ignite replaces H2 row object with its own stub. So, when you ask for a field value, ignite get entry represented by the row and get entry field according to configured mapping (via Queryentity [1] or via annotations [2]). Ignite stores data in serialized way, as every field can be accessed without entire entry\value deserialization. H2 indices also has an Ignite implementation underneath and contains either inline value [3] or reference to value location in-memory, so, they should be very fast and their cost shoudn't be too much.
[1] https://apacheignite.readme.io/docs/indexes#queryentity-based-configuration [2] https://apacheignite.readme.io/docs/indexes#section-annotation-based-configuration [3] https://ignite.apache.org/releases/latest/javadoc/org/apache/ignite/configuration/CacheConfiguration.html#getSqlIndexMaxInlineSize() On Wed, Oct 11, 2017 at 5:37 PM, Ray <[email protected]> wrote: > So when I run a sql in h2 debug console, how does h2 get results from cache > store? > In other words, how does the key/value cache entry to h2 entry mapping > work? > > > > -- > Sent from: http://apache-ignite-users.70518.x6.nabble.com/ > -- Best regards, Andrey V. Mashenkov
