On 10/25/2016 01:12 AM, Jens Alfke wrote:
On Oct 24, 2016, at 1:31 AM, Rowan Worth <row...@dug.com> wrote:
OK, so the entire cache is invalidated when another process updates the DB,
which is what I feared. In this case I'm looking at too many concurrent
updates for caching to add much value.
I’m no expert on the internals, but the article linked to says that this
counter isn’t used in WAL mode. And with the WAL the main database file is
updated less frequently, so the cache might remain valid longer. (Again, I’m
just hypothesizing.)
WAL mode does the same thing - just uses a different mechanism. If some
other process modifies the db and then your process opens a new read
transaction, it discards any cached pages it has in memory.
Dan.
_______________________________________________
sqlite-users mailing list
sqlite-users@mailinglists.sqlite.org
http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users
