# A lot of records running slow on Dreamhost - any suggestions **URL:** \ **Category:** Deployment\ **Created:** [February 20, 2008, 3:57pm UTC](https://www.ruby-forum.com/t/a-lot-of-records-running-slow-on-dreamhost-any-suggestions/129674 "2008-02-20T15:57:50Z")\ **Posts on this page:** 10\ **Page:** 1
**Author:** ![fonziguy](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/f/f05b48/32.png) [@fonziguy](https://www.ruby-forum.com/u/fonziguy)\ **Post date:** [February 20, 2008, 3:57pm UTC](https://www.ruby-forum.com/t/a-lot-of-records-running-slow-on-dreamhost-any-suggestions/129674/1 "2008-02-20T15:57:50Z")
Hi guys, I’m quite new to Ruby on Rails and I am deploying my first app. I have over 50,000 records and each record has a true or false active state. On the homepage I find 50 of the active records and order them randomly. ```auto def index @records = Record.find_all_by_active(true, :order => 'rand()', :limit => 50) end ``` The problem is that my app is running pretty slow and I’m assuming it’s because there’s a lot of records to index through. I have heard about different types of caching but I’m not sure if this is applicable here as I want the records that are displayed to change on each visit. I’m running on Dreamhost & Apache (using Fast CGI). Would appreciate any suggestions or links. Thanks ---
**Author:** ![fonziguy](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/f/f05b48/32.png) [@fonziguy](https://www.ruby-forum.com/u/fonziguy)\ **Post date:** [February 20, 2008, 5:32pm UTC](https://www.ruby-forum.com/t/a-lot-of-records-running-slow-on-dreamhost-any-suggestions/129674/2 "2008-02-20T17:32:48Z")
Did you create an index on the ‘active’ boolean column? This would help a lot if it’s a database performance bottleneck. You could also use page or fragment caching to speed things up further but make sure it has a proper index. Raul ----- Original Message ----- From: “Lee M.” [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) To: [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) Sent: Wednesday, February 20, 2008 6:57 AM Subject: [Rails-deploy] A lot of records running slow on Dreamhost - any suggestions ---
**Author:** ![fonziguy](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/f/f05b48/32.png) [@fonziguy](https://www.ruby-forum.com/u/fonziguy)\ **Post date:** [February 21, 2008, 12:33am UTC](https://www.ruby-forum.com/t/a-lot-of-records-running-slow-on-dreamhost-any-suggestions/129674/3 "2008-02-21T00:33:45Z")
On 2/21/08, Lee M. [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) wrote: > @records = Record.find\_all\_by\_active(true, :order =\> ‘rand()’, > :limit =\> 50) you might want to double check, but I recall something about the rand() function of mySQL being a slow pig. What is the performance like without the rand() ---
**Author:** ![fonziguy](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/f/f05b48/32.png) [@fonziguy](https://www.ruby-forum.com/u/fonziguy)\ **Post date:** [February 21, 2008, 1:11am UTC](https://www.ruby-forum.com/t/a-lot-of-records-running-slow-on-dreamhost-any-suggestions/129674/4 "2008-02-21T01:11:54Z")
@Anthony - Good catch. I forgot about that. But with the LIMIT of 50 it shouldn’t be that big of a deal. Maybe the generated SQL is not what we think it is? I guess you could also try replacing this with a find\_by\_sql statement and see if it performs any better. @records = Record.find\_by\_sql(SELECT \* FROM records WHERE active=1 ORDER BY rand() LIMIT 50") Raul ----- Original Message ----- From: “Anthony R.” [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) To: [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) Sent: Wednesday, February 20, 2008 3:32 PM Subject: [Rails-deploy] Re: A lot of records running slow on Dreamhost - any suggestions ---
**Author:** ![fonziguy](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/f/f05b48/32.png) [@fonziguy](https://www.ruby-forum.com/u/fonziguy)\ **Post date:** [February 21, 2008, 1:34am UTC](https://www.ruby-forum.com/t/a-lot-of-records-running-slow-on-dreamhost-any-suggestions/129674/5 "2008-02-21T01:34:10Z")
Ugh! You sir, are correct! What a drag… Again, great catch. Raul ----- Original Message ----- From: “Anthony R.” [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) To: [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) Sent: Wednesday, February 20, 2008 4:30 PM Subject: [Rails-deploy] Re: A lot of records running slow on Dreamhost - any suggestions ---
**Author:** ![fonziguy](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/f/f05b48/32.png) [@fonziguy](https://www.ruby-forum.com/u/fonziguy)\ **Post date:** [February 21, 2008, 1:31am UTC](https://www.ruby-forum.com/t/a-lot-of-records-running-slow-on-dreamhost-any-suggestions/129674/6 "2008-02-21T01:31:15Z")
But the rand() would need to be performed on the entire dataset before the limit is applied otherwise it wouldn’t be random! Anthony ---
**Author:** ![fonziguy](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/f/f05b48/32.png) [@fonziguy](https://www.ruby-forum.com/u/fonziguy)\ **Post date:** [February 21, 2008, 1:37am UTC](https://www.ruby-forum.com/t/a-lot-of-records-running-slow-on-dreamhost-any-suggestions/129674/7 "2008-02-21T01:37:07Z")
Problem is that it has to do the rand first, then the limit afterwards. I just tried this on a table with around 4 million records: select \* from galleries order by rand() limit 10; I stopped it after 3 minutes. This took .001 seconds on the same table: select \* from galleries order by id limit 10; On an unindexed field, it took 3.5 seconds: select \* from galleries order by thumb\_exists limit 10; Order by rand() really sucks. We try not to use it at all. - Ericson S. [http://www.funadvice.com](http://www.funadvice.com) On Wed, Feb 20, 2008 at 7:30 PM, Anthony R. \< ---
**Author:** ![fonziguy](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/f/f05b48/32.png) [@fonziguy](https://www.ruby-forum.com/u/fonziguy)\ **Post date:** [February 21, 2008, 1:52am UTC](https://www.ruby-forum.com/t/a-lot-of-records-running-slow-on-dreamhost-any-suggestions/129674/8 "2008-02-21T01:52:23Z")
I saw this workaround on the web: SELECT \* FROM table WHERE RAND()\>0.9 ORDER BY RAND() Here’s the site link: [SQL to Select a random row from a database table](http://www.petefreitag.com/item/466.cfm) Raul ----- Original Message ----- From: Ericson S. To: removed\_email\_address@domain.invalid Sent: Wednesday, February 20, 2008 4:36 PM Subject: [Rails-deploy] Re: A lot of records running slow on Dreamhost - any suggestions ---
**Author:** ![fonziguy](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/f/f05b48/32.png) [@fonziguy](https://www.ruby-forum.com/u/fonziguy)\ **Post date:** [February 21, 2008, 6:33am UTC](https://www.ruby-forum.com/t/a-lot-of-records-running-slow-on-dreamhost-any-suggestions/129674/9 "2008-02-21T06:33:41Z")
On 21.2.2008, at 4.05, Anthony R. wrote: > depending on the business requirements for doing a random select > anyway, you may be able to create “randomvalue” column on the table > and then do a select against that. Have a background thread (or a > stored procedure) that then updates all the rows with a random value > every 10 minutes (or every hour?). That way the time consuming part is > not customer facing. > > This depends on why you need a random selection in the first place > ofcourse! Here’s Jan Kneschke’s take on the subject: > **[ORDER BY RAND()](http://jan.kneschke.de/projects/mysql/order-by-rand/)** > > If you read the MySQL manual you might have seen the ORDER BY RAND() to randomize the the rows and using the LIMIT 1 to just take one of the rows. SELECT name FROM random ORDER BY RAND() LIMIT 1; This example works fine and is fast if you only ... //jarkko – Jarkko L. > **[jlaine.net](https://jlaine.net/)** > > Your freedom to carry a gun messes with my freedom to walk around without fear. Your freedom to “stand your ground” with force is at odds with my … > **[Home](https://dotherightthing.com/)** > > do the right thing [http://www.railsecommerce.com](http://www.railsecommerce.com) [http://odesign.fi](http://odesign.fi) ---
**Author:** ![fonziguy](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/f/f05b48/32.png) [@fonziguy](https://www.ruby-forum.com/u/fonziguy)\ **Post date:** [February 21, 2008, 3:06am UTC](https://www.ruby-forum.com/t/a-lot-of-records-running-slow-on-dreamhost-any-suggestions/129674/10 "2008-02-21T03:06:12Z")
depending on the business requirements for doing a random select anyway, you may be able to create “randomvalue” column on the table and then do a select against that. Have a background thread (or a stored procedure) that then updates all the rows with a random value every 10 minutes (or every hour?). That way the time consuming part is not customer facing. This depends on why you need a random selection in the first place ofcourse! Cheers, Anthony R.