Google is deprecating the Master/Slave Datastore used by its cloud platform App Engine in favor of the High-Replication Datastore (HRD), which offers better reliability, the company said [[xref:http://googleappengine.blogspot.co.uk/2012/04/masterslave-datastore-thanks-for-all.html|in blog post on Wednesday. |Google App Engine Blog: Master/Slave Datastore, thanks for all your hard work]]
App Engine allows users to build and host Web applications on Google's infrastructure, and the datastores are used for storage.
The HRD option provides higher availability and better, more predictable performance compared to the Master/Slave Datastore, the blog post said. It was originally announced in January 2011, following a six-month period during which Google struggled with reliability issues.
Although it is massively scalable, the Master/Slave Datastore's availability is dependent on a single datacenter, which when you're serving hundreds of thousands of applications isn't sufficient, Google said.
The HRD addresses that issue by replicating data across multiple data centers. It can withstand multiple datacenter outages and there is no planned downtime, according to Google.
Many upcoming features will only be available on the HRD, according to the blog post.
To make the migration easier, Google has developed a set of migration tools, which can be accessed in the application settings tab of the Administration Console.
However, the Master/Slave datastore isn't going away over night. For a period of three years, Google will use "commercially reasonable efforts" to continue to operate the datastore and to respond to problems "deemed by Google in its discretion to be critical," according to the terms of service.
Send news tips and comments to firstname.lastname@example.org
Join the CIO Australia group on LinkedIn. The group is open to CIOs, IT Directors, COOs, CTOs and senior IT managers.