# Backend Clustering solution, recommendations

**URL:** <https://www.ruby-forum.com/t/backend-clustering-solution-recommendations/131966>\
**Category:** Rails\
**Created:** [March 14, 2008, 8:00pm UTC](https://www.ruby-forum.com/t/backend-clustering-solution-recommendations/131966 "2008-03-14T20:00:52Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![divotdave](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/d/bbe5ce/32.png) [@divotdave](https://www.ruby-forum.com/u/divotdave)\
**Post date:** [March 14, 2008, 8:00pm UTC](https://www.ruby-forum.com/t/backend-clustering-solution-recommendations/131966/1 "2008-03-14T20:00:52Z")

</div>

Hi, I was wondering if anyone could recommend a database clustering  
solution for a massive warehouse. Rails/flex being the main front end.

Talking a few million rows, with queries that may touch a million rows  
every few seconds.

I was giving some thought to a MySQL clustering project like Sequoia.  
[http://sequoia.continuent.org/HomePage](http://sequoia.continuent.org/HomePage)

Or possibly just hiring a consulting firm to come in.

We’ve done some good size projects, but this will be the biggest we’ve  
done to date.

– Thanks

---

<div class="post-metadata">

**Author:** ![divotdave](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/d/bbe5ce/32.png) [@divotdave](https://www.ruby-forum.com/u/divotdave)\
**Post date:** [March 14, 2008, 8:20pm UTC](https://www.ruby-forum.com/t/backend-clustering-solution-recommendations/131966/2 "2008-03-14T20:20:45Z")

</div>

> Hi, I was wondering if anyone could recommend a database clustering  
> solution for a massive warehouse. Rails/flex being the main front end.
> 
> Talking a few million rows, with queries that may touch a million rows  
> every few seconds.
> 
> I was giving some thought to a MySQL clustering project like Sequoia.  
> [http://sequoia.continuent.org/HomePage](http://sequoia.continuent.org/HomePage)

We tried continuent a year or two ago and gave up. And our database was  
small (couple hundred megs). We had three nodes and had problems with a  
node crashing bringing the other nodes down. With the result being we  
had  
to power cycle each of the boxes. We never made it to production.

That said, it was about two years ago. Hopefully it’s improved, but  
I’ve  
been soured by the experience. Also, that wasn’t sequoia, it was their  
commercial offering. No idea what difference that makes.

In our case we ended up just using a single master with multiple slaves  
and the mysql\_replication\_adapter to spread the load.

-philip

---

<div class="post-metadata">

**Author:** ![divotdave](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/d/bbe5ce/32.png) [@divotdave](https://www.ruby-forum.com/u/divotdave)\
**Post date:** [March 21, 2008, 3:05am UTC](https://www.ruby-forum.com/t/backend-clustering-solution-recommendations/131966/3 "2008-03-21T03:05:39Z")

</div>

That kind of replication is more than you really want to have on a  
simple  
master-master or master-slave replication. Your better off bringing in  
someone with experience with that scale of DB, and using whatever they  
recommend to you.

Rob K.

> **[SoftServ | Services](https://softserv.scientist.com/)**
>
> SoftServ is a software development consultancy that turns software ideas into digital realities.