Hello Christo and thank you for your feedback. For the moment we have a Relational Database that is not distributed or partitioned. We have given some thought for configuring at least some database tables as cache store to Ignite and it not something that we exclude. Just for the moment due to some Network I/O performance issues of the Cloud platform that will host the solution we cannot say much.
However, there is a strong chance at least one or two tables involved in the flow I described to be loaded into cache. For this reason I would like to ask the following: 1) Which is the best policy for loading into cache a table of about ~1.000.000 rows and use ANSI SQL to query it later? This table will be incremented periodically (after an ETL procedure) utill it will reach a final upper limit of about ~3.000.000 rows. So I would like the optimal way to stream it initially into cache on startup of a future deployment and update it after each ETL procedure. 2) If we use MapReduce & ForkJoin procedure how can we combine it with Affinity? There are examples for Distributed Closures but I do not see any for ComputeTask/ComputeJobAdapter etc. Ideally each job, should perform ANSI SQL query on the table that will be loaded and maintained in cache but on the rows that it keeps locally. -- View this message in context: http://apache-ignite-users.70518.x6.nabble.com/Distributed-Closures-VS-Executor-Service-tp11192p11653.html Sent from the Apache Ignite Users mailing list archive at Nabble.com.