Howdy! I perused the documentation and wiki entries, and couldn’t
find an answer to this. If it’s obvious (and in the documentation),
please feel free to tell me to RTFM and point me toward the page J We have a multithreaded application that uses a sqlite3
database for asset tracking. I’m having some performance issues, and was
wondering how best to proceed. We’re storing arbitrary blobs in the
database, and retrieving them in a standard manner (open, prep statement,
sqlite3_step). Our problem is that the sqlite3_step often takes up to 1200
milliseconds to retrieve a blob. Granted, the blobs may be large, but 1200
millis seems excessive. I’d like to narrow down what’s causing the
problem. Our queries, made by a person very familiar with SQL and optimizing
SQL, seem sane to me. How do I narrow down if the time sink is: 1) The query
itself 2) The recovery
of the blob 3) File
collision or locking Any pointers toward documents, examples, or hints on how to
narrow down the scope of the performance issue would be greatly appreciated. Thanks! Mitch
|
- [sqlite] sqlit3_step question Evans, Mitchell