[Okta Webinar] Learn how to a build a cloud-first strategyRegister Now

x
  • Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 667
  • Last Modified:

MySQL in Windows cluster

I am a MS SQL DBA/developer  - zero expereince with MYSQL. I am trying to ascertain whether the follwing is posslbe and what the best approach would be.

We have a 2 box Windows (2k3)  server cluster with attached SAN.  We are currently runnnig MSSQL 2000 Enterprise (in a clustered configuration)
We are investigating the use of a new application that only supports a MYSQL database.

We need a high availabilty model (i.e. a cluster) - what are the options for clustering MYSQL?
Preferably, we'd like to find an "enterprise" version that can be clustered using MSCS - is there such a thing?

If not, then what s the best approach, bearing in mind that we already have the Windows Server cluster

From my intial reading it seems that MYSQL has it's own clustering model (which seems to use replication??)
0
OctaviaIS
Asked:
OctaviaIS
2 Solutions
 
Kim RyanIT ConsultantCommented:
Lots of detail here. No specific reference to cluster model. If you wade through the white paper or search the forum, you should get more detail.
http://www.mysql.com/products/database/cluster/
0
 
arpoodleCommented:
native MySQL Clustering, as available from 5.0, is only available in Linux.  You can perform 'kind-of' clustering using circular replication between windows boxes.

Clustering in mySQL is a shared nothing architecture that uses it's own storage engine.  in 5.0 it's an in-memory data storage engine.  5.1 sees that moving to in-memory with a disk absed commit. Clustering uses network interconnects between storage nodes, and will automatically detect when nodes come online(5.1 and later) and offline.

replication is simpler, in that it is a binlog of all update,insert and deletes performed against a database, and the slave database pulls that info and 'replicates' the queries.  Circular replication sees a master replicate to a slave, that then replicates back to the master, so you have (in effect) a 2 box cluster.  Replication doesn't have any way of knowing that the command has been completed on the slave.  Clustering looks for confirmation.

MySQL Clustering is still very new, but well worth looking at, but as I said, it's only available with the linux distribution as far as I know.

There is a company (continuum??) that provides a 3rd party, disk based MySQL clustering solution, but I know very little about it.

a

0

Featured Post

Concerto Cloud for Software Providers & ISVs

Can Concerto Cloud Services help you focus on evolving your application offerings, while delivering the best cloud experience to your customers? From DevOps to revenue models and customer support, the answer is yes!

Learn how Concerto can help you.

Tackle projects and never again get stuck behind a technical roadblock.
Join Now