Hi Lawrence, I appreciate your remarks. However database engines cache their table/views to support sequential accessing within a set. With a good accessing scheme and with enough cache memory you will have all your small tables in memory.
So the simplest thing is let the DBMS do its thing. The good ones will cope quite happily. You then have the advantage that the app can grow without program changes. It is a long time since I delved into DBMS internals (25 years) but I cannot see that they will have changed from what I have said above, however I am sure to be corrected if I am wrong. ;-) Cheers, phil -----Original Message----- From: Lawrence D'Oliveiro [mailto:l...@geek-central.gen.new_zealand] Sent: Sunday, 31 May 2009 11:21 p.m. To: python-list@python.org Subject: Re: which database is suitable for small applications In message <52801358-c037-458d-8857- a78c2d881...@z16g2000prd.googlegroups.com>, Ankit wrote: > If your application does not require you to add very heavy data then > you can also use flat files to do your stuff. > Its always a better to use Databases ... It's not always better to use databases. By definition, data for a “small” application is bound to fit entirely in RAM. So certainly the simplest thing to do is read a whole data file in at the start, and write the entire updated data structure back out again at the end. -- http://mail.python.org/mailman/listinfo/python-list
