I am thinking it might be worth setting up a demo site with > 1 million records just for interest. We are already doing that using the InfoBright Community Edition (ICE) plugin for MySQL for very complex OLAP queries.
-Owen On Fri, Apr 9, 2010 at 1:25 PM, kevinpfromnm <[email protected]> wrote: > would have to agree in general, db is not your bottleneck. view > rendering is going to be what sets how much hardware you need which is > all about how many hits per second. > > On Apr 9, 8:23 am, Owen Dall <[email protected]> wrote: > > Agreed. Indexing is the key. It is not how many records, but how > > complicated are the joins with related records. Even with that, Hobo 1.0 > > provides good indexing out of the box. > > > > BTW Matt wrote the enhanced Hobo indexing for 1.0. He has a good > tutorial > > on customizing indexes in the "Rapid Rails with Hobo" book: > > > > Tutorial 25 – Using Hobo Database Index Generation > > > > -Owen > > > > > > > > On Fri, Apr 9, 2010 at 10:17 AM, Matt Jones <[email protected]> wrote: > > > > > On Apr 9, 2010, at 10:05 AM, [email protected] wrote: > > > > > Well, I just went through the Agility tutorial - all I can say is WOW! > > >> This is truly good stuff, and I can see that I've got quite a bit to > > >> learn. > > > > >> I'm curious as to how scalable this might be ... lets say that I have > > >> 1000 users, and each user has 130 tasks of their own, along with 300 > > >> contacts of their own plus several other tables of data where each > > >> user may have lets say 100 records. You are already up to over 500 > > >> records per user, and over 500,000 records total ... my question is > > >> whether or not using rails + hobo + apache + mysql is a good, > > >> scalable, and fast performing solution for such an application ... or > > >> would some other solution be better? > > > > >> I realize that not all users will be accessing their data at the same > > >> time, but how long will it take to retrieve each users set of 300 > > >> contacts from a table that has 300,000 entries? > > > > > Not long, if the table is properly indexed. Fortunately, Hobo tries > pretty > > > hard to ensure that you get the indexes likely to help - if you declare > > > Contact belongs_to User, the migration generator will automatically add > an > > > index on user_id. > > > > > As far as the DB goes, MySQL apparently scales well past millions of > > > records, and the Postgres crowd does even more AFAIK. Actually, the > biggest > > > problem in database scaling comes from when you start having > cross-relations > > > between the data. The basic setup you've described (without links > between > > > the users) is highly susceptible to advanced techniques like shardin. > > > > > Hope this helps! > > > > > --Matt Jones > > > > > -- > > > You received this message because you are subscribed to the Google > Groups > > > "Hobo Users" group. > > > To post to this group, send email to [email protected]. > > > To unsubscribe from this group, send email to > > > [email protected]<hobousers%[email protected]> > <hobousers%[email protected]<hobousers%[email protected]> > > > > > . > > > For more options, visit this group at > > >http://groups.google.com/group/hobousers?hl=en. > > > > -- > > Thanks, > > - Owen > > > > Owen Dall, Chief Systems Architect > > Barquin International > > Cell: 410-991-0811 > > -- > You received this message because you are subscribed to the Google Groups > "Hobo Users" group. > To post to this group, send email to [email protected]. > To unsubscribe from this group, send email to > [email protected]<hobousers%[email protected]> > . > For more options, visit this group at > http://groups.google.com/group/hobousers?hl=en. > > -- Thanks, - Owen Owen Dall, Chief Systems Architect Barquin International Cell: 410-991-0811 -- You received this message because you are subscribed to the Google Groups "Hobo 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/hobousers?hl=en.
