I have recently downloaded infobright and would like to use it for a data warehousing project I’m working on.
I chose it mainly because of the extreme data compression advantage, but I don’t know if it’s capable of serving my purposes due to the potential limits on the number of concurrent queries.
I need to use it to store data which is greater than 10 TB in compressed InnoDB tables. In infobright, it will be much smaller (obviously).
This project will involve thousands of tables, 25% of which will be infobright (these are the largest), 75% of which will be myISAM or InnoDB.
I will need to do joins between infobright and myIsam or InnoDB tables. I realize that I will take a hit on performance with this due to the slowness encountered when joining brighthouse tables with other table types. There’s nothing I can do about this; however, as I need the other table types due to InfoBright community edition not supporting SQL DML.
Users of this system will be consistently querying the database, resulting in potentially hundreds of queries at a time. Many of these queries will return 1000 rows or less.
The computer I plan on using has a 2.5 Ghz Intel Quadcore Processor with 4 GB of ram.
Will the brighthouse engine be able to handle this?
Do I need significantly more memory?
Any help with this would be greatly appreciated.
