[RESOLVED] Experience with Database (MySQL) Replication?

dalecosp

I'm looking at a need for, I believe, database replication.

Mysql Manual wrote:
“Analytics - live data can be created on the master, while the analysis of the information can take place on the slave without affecting the performance of the master.”

Which matches this business need to a "T", near as I can figure.

What I'm concerned about is the impact this will have on the "master" (which is the production web environment):

Mysql Manual wrote:
Replication between servers in MySQL is based on the binary logging mechanism. The MySQL instance operating as the master (the source of the database changes) writes updates and changes as “events” to the binary log.

Does anyone have much experience with this? MySQL is already something of a bottleneck on the production box; I'd prefer not to do anything to make it worse.... :eek:

NogDog

Only experience I have is a teensy bit with Postgres. I didn't notice any performance hits, but then the vast majority of our DB operations are selects, whereas I expect most replication activities are triggered by inserts/updates? It actually helped our performance a bit, as we redirected one of the heavy request sources to the slave db server, so that all those selects were split up between the master and the slave. 🙂

Weedpacket

In the same situation as NogDog, here. My guess though would be that, as the master is writing to the binlog anyway for purposes of crash recovery and the slaves are reading from that, there wouldn't be much reason for the former to suffer.

Unless of course both dbmss are on the same machine and have to fight each other for CPU cycles 😃

dalecosp

Weedpacket;11030753 wrote:
In the same situation as NogDog, here. My guess though would be that, as the master is writing to the binlog anyway for purposes of crash recovery

Ooh, I didn't know that.

I hope, by this point, I know better than to ask, "are you sure"? So, I might just beg for a link? 😃

dalecosp

NVM ... Google's my friend, too 🙂