Re: JVM tuning parameters

We are testing and also the production data would be so much. -- Sent from: http://apache-ignite-users.70518.x6.nabble.com/

Re: When I use Ignite create table ddl，can I set the db schema instead of the schema name "PUBLIC"？

All, I need to create a custom schema and then use SQL to create tables in that schema. The 2.7 docs imply I can. Is this now possible? If so, how? I have spent many hours trying. Questions: How do I create a custom schema? I attempted to add the custom schemas to the xml and it

JDBC Thin Client Set Streaming On/Off in node.js app

Hi all, I currently have an ETL application that parses raw logs and builds batch inserts that load our Oracle table. We're trying to build out our Apache Ignite environment that will load a subset of the data. Using the IgniteJdbcThinDriver I wrote a simple node.js program that connects to our

Re: Stream continuous Data from Sql Server to ignite

Hello! This is impossible in general, since databases do not have uniform push API. Moreover it is tricky WRT transactional changes. It may be possible for specific database (triggers, etc), but you will have to implement the bulk of logic, or maybe use a 3rd party solution. Regards, -- Ilya

JVM tuning parameters

Hi Team, We are installing ignite on kubernetes environment with native persistence enabled. When we try to create around 400 tables using the sqlline end point , the pods are restarting after creating 200 tables with jvm heap error so we have increased the java heap size from 1GB to 2GB and

Re: JVM tuning parameters

Hello! Ignite has around 20M heap overhead per table, this means 6G for 300 tables. Why do you need so many tables? Regards, -- Ilya Kasnacheev чт, 28 мар. 2019 г. в 14:47, ashfaq : > Hi Team, > > We are installing ignite on kubernetes environment with native persistence > enabled. When we

Re: Stream continuous Data from Sql Server to ignite

As per my knowledge by enabling writeThrough any updates to the cache will be reflected to its underlying RDBMS tables. Whereas we want the other way, updates in RDBMS table should be reflected in cache. Thanks, Austin -- Sent from: http://apache-ignite-users.70518.x6.nabble.com/

Re: Cache Join

Hello! Do you really need such huge number of fields to join? Isn't it easier to introduce an affinity field? Anyway, you seem to create an index with a huge number of fields. Consider only keeping fields which are necessary for performing that join. You might even need to boil it down to some

Re: xa transaction manager and ignite

Hello! Unfortunately I have no idea here. Maybe anyone more knowing will chime in. Regards, -- Ilya Kasnacheev чт, 28 мар. 2019 г. в 00:37, Scott Cote : > Yes I did. It is in fact the basis for my solution  > > > > Right now, I’m doing some fakery …. > > > > Taking the Spring

Re: Support User defined aggregate function ?

Currently, they are not supported, but you could ask about it on dev-list and maybe somebody implement it. Or propose PR with UDAF support Also use-cases for UDAF usages will be helpful -- Sent from: http://apache-ignite-users.70518.x6.nabble.com/