Comments on: Harddrivers are cheap, CPU’s aren’t; cache your data, fool! http://blog.jorygeerts.com/2007/01/06/harddrivers-are-cheap-cpus-arent-cache-your-data-fool/ Just another WordPress weblog, or something like that Sun, 16 May 2010 16:52:11 +0000 hourly 1 http://wordpress.org/?v=3.0.1 By: Jory http://blog.jorygeerts.com/2007/01/06/harddrivers-are-cheap-cpus-arent-cache-your-data-fool/comment-page-1/#comment-2543 Jory Sun, 07 Jan 2007 13:16:57 +0000 http://blog.jorygeerts.com/2007/01/06/harddrivers-are-cheap-cpus-arent-cache-your-data-fool/#comment-2543 Man, your so much better at explaining stuff that I am.. thats exactly what I meant. :P Man, your so much better at explaining stuff that I am.. thats exactly what I meant. :P

]]> By: Jeremy Privett http://blog.jorygeerts.com/2007/01/06/harddrivers-are-cheap-cpus-arent-cache-your-data-fool/comment-page-1/#comment-2542 Jeremy Privett Sun, 07 Jan 2007 01:12:59 +0000 http://blog.jorygeerts.com/2007/01/06/harddrivers-are-cheap-cpus-arent-cache-your-data-fool/#comment-2542 Whenever I'm working on an application that needs data parsed like what you described above, I always add one additional field to the database table to hold the unparsed data. And the only time this data is presented is to an editor. When the edit is complete, the unparsed data gets updated and the parsed data is overwritten with the new reflection of the unparsed. Does this take up more disk space? Yes. Who cares? Not me. When you start to talk about a high volume of visitors to the application, scalability is important. And this is the most scalable solution. You parse once when data is submitted, and never have to parse again, unless it's edited. Simple logic, really. Don't waste extra CPU cycles on unnecessary parsing of what is more often than not static data. Whenever I’m working on an application that needs data parsed like what you described above, I always add one additional field to the database table to hold the unparsed data. And the only time this data is presented is to an editor. When the edit is complete, the unparsed data gets updated and the parsed data is overwritten with the new reflection of the unparsed.

Does this take up more disk space? Yes. Who cares? Not me. When you start to talk about a high volume of visitors to the application, scalability is important. And this is the most scalable solution. You parse once when data is submitted, and never have to parse again, unless it’s edited. Simple logic, really.

Don’t waste extra CPU cycles on unnecessary parsing of what is more often than not static data.

]]>