In Memory Data Sources

Questions and answers on designing your Servoy solutions, database modelling and other 'how do I do this' that don't fit in any of the other categories

In Memory Data Sources

Postby swingman » Sat Mar 16, 2019 7:50 pm

Hi all,

Is the actual data loaded into a 'In Memory Data Source' shared by all clients, or can you load different data into each client?

I have a complicated query that will only load a tiny subset of data at the time to stay fast and responsive, so different clients may look at different lists of records simultaneously...
Christian Batchelor
Certified Servoy Developer
Batchelor Associates Ltd, London, UK
http://www.batchelorassociates.co.uk

http://www.postgresql.org - The world's most advanced open source database.
User avatar
swingman
 
Posts: 1472
Joined: Wed Oct 01, 2003 10:20 am
Location: London

Re: In Memory Data Sources

Postby rvanderburg » Sat Mar 16, 2019 10:09 pm

Its per client
rvanderburg
Site Admin
 
Posts: 78
Joined: Wed May 04, 2011 10:28 am

Re: In Memory Data Sources

Postby swingman » Mon Mar 18, 2019 12:18 am

Great. Thanks for confirming.

I have now found more info here:

UX Pattern: Using In Memory Data in Servoy to Accelerate delivery
https://www.youtube.com/watch?v=_FS92nqXibk

I had missed the fact that Servoy has an in-memory database that you can connect to like any other database if you want to share in-memory data between clients....

I think the client will be impressed ;-) by what I have built combining postgreSQL windowing functions and Servoy in-Memory data...
It is crazy fast even though its doing running totals on up to 50,000 records...
Christian Batchelor
Certified Servoy Developer
Batchelor Associates Ltd, London, UK
http://www.batchelorassociates.co.uk

http://www.postgresql.org - The world's most advanced open source database.
User avatar
swingman
 
Posts: 1472
Joined: Wed Oct 01, 2003 10:20 am
Location: London

Re: In Memory Data Sources

Postby sean » Mon Mar 18, 2019 3:08 pm

Hi Christian,

( Thanks for the webinar plug ;-) )

Yes you can create a mem table which is shared. It functions the same way as other shared data sources, using HSQLDB.
Only difference is that the schema must be recreated every time the server starts and then you probably want to seed it with data, usually done with a script, either by first-access lazy load, or by batch processor.

In developer you can have DBI files, and other data-bound constructs (forms, valuelists, etc) which will generate warnings on start-up, but then you can sync to DBI files to resolve and auto build the schema as well.
Software Engineer
Servoy USA
sean
 
Posts: 370
Joined: Mon May 21, 2007 6:26 pm
Location: USA


Return to Programming with Servoy

Who is online

Users browsing this forum: Bing [Bot] and 13 guests