> > I just tried (hadn't noticed that option before) to go from > 2000 to 4000 and 8000, without noticing any difference. I > might try next week to raise the page size to 50k and see if > it makes a difference? >
On the presumption the Sqlite allocates new pages on the page boundary I would suggest you use multiples of the file system page size. I believe (but I could be wrong) both Linux and Windows NTFS defaults to 4096 bytes. If you chose 4,500 bytes in this scenario, for every call to retrieve one Sqlite page, it would require two calls to the OS. For whats it worth, when I did some performance testing a few years ago I recall there was no significant differenence using larger page sizes on a standard desktop machine but your mileage may vary. On a related note I do remember that when I was testing large reads (1M, 10M , 100M) using a single 'fread' call compiled in MSVC 2005 on Windows XP SP2 it had an interesting side effect in the fact that it performed the function by calling a lower level API multiple times with a size of 65,355 bytes, regardless of the original size requested in 'fread'. I believe this can be shown using the file system tools from SysInternals. Therefore I suspect there will be little-to-no benefit of page sizes greater than 64kiB on Windows XP. Nick ******************************************************************** This email and any attachments are confidential to the intended recipient and may also be privileged. If you are not the intended recipient please delete it from your system and notify the sender. You should not copy it or use it for any purpose nor disclose or distribute its contents to any other person. ******************************************************************** ----------------------------------------------------------------------------- To unsubscribe, send email to [EMAIL PROTECTED] -----------------------------------------------------------------------------
