diff options
Diffstat (limited to 'content/posts/2004-10-09-sqlite-db-locking-a-problem.html')
| -rw-r--r-- | content/posts/2004-10-09-sqlite-db-locking-a-problem.html | 24 |
1 files changed, 24 insertions, 0 deletions
diff --git a/content/posts/2004-10-09-sqlite-db-locking-a-problem.html b/content/posts/2004-10-09-sqlite-db-locking-a-problem.html new file mode 100644 index 0000000..5c15e44 --- /dev/null +++ b/content/posts/2004-10-09-sqlite-db-locking-a-problem.html @@ -0,0 +1,24 @@ +--- +date: "2004-10-09T22:22:38Z" +title: SQLite DB-Locking a Problem? +--- + +<p> +I've been playing with <a href='http://sqlite.org/'>SQLite</a> for the last couple of days. I'm trying to figure out if it's a suitable replacement for the ad-hoc storage format I've got for <a href='http://raggle.org/'>Raggle</a>. <a href='http://sqlite.org/'>SQLite</a> has a lot of benefits: it's fast, small (the , and free (public domain). It supports sub-selects, atomic transactions, and there's a complete set of <a href='http://sqlite-ruby.sf.net/'>Ruby bindings</a> (which are available as a <a href='http://rubygems.rubyforge.org/'>Gem</a>, but not as a <a href='http://debian.org/'>Debian</a> package... go figure). Of course all this goodness doesn't come without <a href='http://www.sqlite.org/omitted.html'>caveats</a>. Specifically, <a href='http://www.sqlite.org/faq.html#q7'>here's</a> the one that might be a problem for me: +</p> + +<blockquote cite='http://www.sqlite.org/faq.html#q7'> +<p> +Locking in SQLite is very course-grained. SQLite locks the entire database. Big database servers (PostgreSQL, Oracle, etc.) generally have finer grained locking, such as locking on a single table or a single row within a table. If you have a massively parallel database application, you should consider using a big database server instead of SQLite. +</p> + +<p> +Source: <cite><a href='http://www.sqlite.org/faq.html'>SQLite <acronym +title='Frequently Asked Questions'>FAQ</acronym></a></cite> +</p> +</blockquote> + +<p> +<a href='http://raggle.org/'>Raggle</a> isn't a "massively parallel database application", but it can have up to N threads (where N is the number of feeds a user is subscribed to) attempting to write to the feed list simultaneously. I can probably queue database inserts and limit the threads to <code>SELECT</code>ing from their respective tables, but that smacks of hackery, which is what I was trying to avoid in the first place. I guess it's still a better solution than what <a href='http://raggle.org/'>Raggle</a> does right now. Ah well, <em>C'est la vie</em>. +</p> + |