Hi Riccardo, It depends on how you are asking for the record which is not in Ignite cache. If you are doing this using *get* method of Ignite cache API it will load this record from DB (if it's not already in Ignite cache), put it into in-memory cache and return it to you.
But if you are using Ignite SQL and trying to obtain record using something like this: *select * from my-cache where id=123 , * it will return you nothing if such specific record wasn't already loaded into cache. Igor On Mon, Dec 5, 2016 at 8:14 AM, Riccardo Iacomini < [email protected]> wrote: > Hi Val, > I have a related question regarding SQL queries. Does it mean that it > cannot be used as cache layer for an operational DB? What if I ask for a > record which is not currently in the cache, but is stored in the > persistence DB? Will a get an empty result set or Ignite will ask it to the > DB? > > Thank you in advance for the response. I looks like a silly question but I > would like to be sure I got the use case for the SQL case. > > > > Riccardo Iacomini > > > *RDSLab* > > On Fri, Dec 2, 2016 at 6:35 PM, vkulichenko <[email protected] > > wrote: > >> Hi Riccardo, >> >> If you are working with SQL queries, then you have to load all required >> data >> into Ignite in advance. Ignite will not go to Cassandra in this case and >> will return the result based only on entries that are already in caches. >> >> However, if you use key-value API, both read-through and write-through are >> available. I.e. when you read an entry using get() method providing a key, >> Ignite will load value for this key from Cassandra in case it's not in the >> cache yet. >> >> -Val >> >> >> >> -- >> View this message in context: http://apache-ignite-users.705 >> 18.x6.nabble.com/Cassandra-cache-info-tp9370p9372.html >> Sent from the Apache Ignite Users mailing list archive at Nabble.com. >> > >
