Re: Removing SQLite sessions from the default distribution
| From: | George Schlossnagle | Date: | Tue, 01 Jul 2003 20:46:31 +0000 |
| Subject: | Re: Removing SQLite sessions from the default distribution | ||
| References: | 1 | Groups: | php.internals |
| Request: | Send a blank email to internals+get-3114@lists.php.net to get a copy of this message | ||
On Tuesday, July 1, 2003, at 04:19 PM, Sterling Hughes wrote:
Sure. But what Wez said was that it only affected you when you got 150 req/s, which is not true. And I am being honest. Over the period of 25000 requests, it is likely that you will have *more* than 300 sessions, not less. This would further prove my argument. 300 Unique users in 25000 requests to a PHP script is not an unreasonable benchmark, it favors sqlite more than it should.I didn't see any regression based on the number of distinct sessions. In fact, give the nature of the sqlite locking mechanism, I would suspect little change with the number of distinct sessions.
Right. Percentages are nifty like that. :)Oh, and you can't have a '300% slowdown' it's a 66% slowdown. But that's semantics.Yeah, its a 300% speedup to use files. :)