We had a speed increase because we had a lot of text searches in the old system, all going through text fields where the same values were repeated over and over. Its therefore a lot faster to search a much smaller table, where the text fields are unique, and find the value once, then use an ID comparison, being much faster to match integers than text.
On Sun, Mar 8, 2009 at 6:58 AM, Lawrence Krubner <[email protected]>wrote: > > > > On Mar 7, 2:06 am, Gareth McCumskey <[email protected]> wrote: > > Greetings all, > > > > We have recently released a project we have been working on for some > months > > now as an Alpha version and while we have focussed primarily on bug > fixing > > as well as feature completion for the next Alpha release coming up in a > > week, I can't help but notice something disconcerting. > > > > The project we have developed is a replacement of an existing product. > The > > previous version, coded before my time at the company, is old, procedural > > and uses a very inefficient, un-normalised database structure. > > > > For our new version, we decided to use symfony for maintainability > reasons > > as well as the fact that this version will be a lot more complex than its > > predecessor so symfony's ability to simplify the development helps us > > immensely. > > > > The problem I have noticed is that the new symfony version seems to be > > performing ... well ... badly. Loading pages on the new version takes a > lot > > longer, talkin 10-50 times longer than the previous version. I went so > far > > as to view the development logs and manually run SQL queries on our new > > normalised database schema vs the old version un-normalised version and > the > > new schema performs batter by a factor of 100x so I know that it is > > definitely not the database slowing things down. > > I'm surprised you got a speed boost by normalizing the database. It is > often the other way around. The perfectly normalized database tends to > require a lot of JOIN statements. A small degree of de-normalization > can greatly improve performance. Of course, the great risk of de- > normalization is that you are storing redundant data, and you may > eventually end up with a situation where data in table A is different > than data in table B, for a field that is suppose to hold identical > data. > > It's rare to hear of a speed boost coming from normalizing. > > > > > > > --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "symfony users" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/symfony-users?hl=en -~----------~----~----~----~------~----~------~--~---
