Hi Noel! Currently we use original version of H2 database, it is just a Maven dependency in our project. Time ago I committed some improvements for H2 database (I'm still on the committers list). I think the page says about that, so it is done already :)
In general of course we prefer contributing back to original projects instead of forking. Sergi On Tuesday, January 27, 2015 at 8:11:25 PM UTC+3, Noel Grandin wrote: > > Hi sergi > > Nice to see h2 being used for something like that! Awesome! > > I see on that page that grid gain engineers have patched h2. Have those > patches been contributed back? > > Regards, Noel. > > On Tuesday, 27 January 2015, Sergi Vladykin <[email protected] > <javascript:>> wrote: > >> Hi, >> >> Apparently H2 is used as SQL engine in GridGain >> <http://www.gridgain.com/> (a similar to Hazelcast open source grid) to >> query entries in distributed caches. >> Now GridGain is donated to Apache Software Foundation under the name of >> Apache >> Ignite <http://ignite.incubator.apache.org/> and as part of this >> transition we are working on complete SQL support using H2. >> How it works right now you can check here: >> http://doc.gridgain.org/latest/SQL%2C+Scan%2C+And+Full+Text+Queries >> So stay tuned! >> >> BTW, Contributions to Apache Ignite project are welcome! >> >> Sergi >> >> >> On Tuesday, January 27, 2015 at 3:07:57 PM UTC+3, wener wrote: >>> >>> Any plan to add distributability for H2 over hazelcast ? >>> >>> Like OrientDB + Hazelcast >>> Here is a webinar talk about this orientdb-hazelcast- >>> memory-distributed-graph-database >>> <http://hazelcast.com/resources/orientdb-hazelcast-memory-distributed-graph-database/> >>> . >>> >>> Here is what I googled. https://groups.google.com/forum/#!topic/ >>> hazelcast/-qkG8AVM1GA >>> >>> There is no easier answer to this one unfortunately. I am however a big >>>> fan of H2 Db and have spent some time looking at doing exactly what you >>>> have described. Here is the basic approach, which would take a couple of >>>> passes to get right: >>>> >>>> 1. Each node instance has a copy of the H2 database engine. >>>> Relational data structure are stored on the grid along with cache data >>>> 2. Modify the H2 source, injecting Hazelcast Map objects and List >>>> objects where appropriate >>>> 3. Modify thread synchronization related to relational data >>>> structure access and allow Hazelcast to manage >>>> 4. Look at opportunities in the H2 code to use the distributed >>>> executor service on the grid for parallel processing >>>> 5. The trick is to make sure you are processing data on the grid >>>> and mot pulling large data sets back to an individual node too soon. >>>> >>>> H2 does provide the ability to plug-in your own file system fairly >>>> easily, however I don't think you would see any benefit moving the >>>> in-memory file system to a distributed in-memory file system without >>>> addressing concurrency issues at the data structure level. >>>> >>>> >>>> I think it may all be worth doing though :) >>>> >>>> Regards, >>>> Eric >>>> >>> >>> H2 dose has master-slave-abiliy , but it's very limited. >>> >> -- >> You received this message because you are subscribed to the Google Groups >> "H2 Database" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to [email protected]. >> To post to this group, send email to [email protected]. >> Visit this group at http://groups.google.com/group/h2-database. >> For more options, visit https://groups.google.com/d/optout. >> > -- You received this message because you are subscribed to the Google Groups "H2 Database" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/h2-database. For more options, visit https://groups.google.com/d/optout.
