arrow_back Build a Queue Based Concurrent Task Processor (using Python)
Reporting Using MySQL Multi-Source Replication
Submitted by Vishnu H Rao (@vishnuhr) on Friday, 12 April 2013
Storage and Databases
How can we do real time reporting using MySQL when data is spread across different MySQL instances?
We can achieve this using Multi-Source replication.
Multi-Source replication is used for consolidating data from different MySQL instances.
There are two major use cases for Multi-Source replication:
(1) Backup at a single place for data spread across different MySQL instances. (2) Real Time Reporting - the capability to do complex joins across tables which were otherwise present in different Instances - say a join across Orders & Invoices Tables.
This session would present different ways of achieving Multi-Source replication in MySQL 5.5, the pros & cons of each approach, solutions to failure scenarios i.e. what if a MySQL instance we were replicating from fails, how would failover take place.
This session would also talk about the Relay Log & its power to make failover very easy.
I work for Flipkart - India's largest e-commerce company.
I am part of the Flipkart Data Platform team - whose charter is to take care of Flipkart's data infrastructure needs.
I also serve as Chief Surgeon at the Flipkart M*A*S*H - MySQL Army Surgical Hospital - handling MySQL emergencies.
I also love doodling - inspired from Google - i maintain my own doodle blog @ doodle-vishnu.blogspot.in
My favorite t.v series are M*A*S*H & Remington Steele.
A Sharks & Ducks fan - i love the NHL.