Hi everyone,<br><br>I've been following the topic of storing large amounts of data in Mnesia.<br>The overall problem seems to be the fragmentation of free space when deleting records, correct?<br><br>My question is: Can this issue be avoided if records where not physically deleted but only _marked_ as deleted by a field in the record?
<br><br>Alternatively: Would it be possible to have a timer job which would take the node offline, repair the table and put it online again?<br>In a clustered environment this could be done in a round-robin kind of fashion with no visible downtime to users.
<br><br><br>I would really like to use Mnesia in a large web application but the size limitation scares me.<br>If anyone can think of a workaround I would very much like to know...<br><br>- Mikkel<br>