Google Cloud SQL: Relational Database on the Cloud

Ajay Ohri
Aug. 13 2012, 11:30AM EDT

In our concluding part of our series on Google Cloud APIs we cover Google Cloud SQL. We have covered Google Prediction API, Storage API, and BigQuery API earlier.

More than 10,000 developers have joined Google's RDBMS in the cloud , the Google Cloud SQL. This is because just for a dollar per month you can get started with a cloud hosted MySQL database. Support for Google Cloud is available through Java and Python , including an Eclipse Plugin and also a command line tool (through the Squirrel SQL tool). However one of the easiest way to access Cloud SQL is through the SQL Prompt tab in Google APIs console.

Pricing (introduced in June 2012)  is both package based and pay per use based. Data Limits are quite reasonable-Both the maximum request size and maximum response size are limited to 16 mb and time out limits are as follows.

  • All database requests must finish within the HTTP request timer, around 60 seconds.
  • Offline requests like cron tasks have a time limit of 10 minutes.
  • Backend requests to Google Cloud SQL have a time limit of 10 minutes.

While the Google Cloud SQL is highly recommended if you need RDBMS type data and are using Google App Engine platform, what is relatively unclear is how someone with an existing ,PostgreSQL, MySQL, or Oracle RDBMS can transition to Google Cloud SQL. The lack of clarity in how to transition for existing RDBMS users who are new to Google Apps, is one big thing that holds it back from earning in the lucrative database market.

Also while players like Oracle and SAP have embraced in-database analytics (both through the R language), Google who has been one of the earliest supporter and founder of R, has not made it in-database analytics plans clear for the Cloud SQL.

Make no mistake- we love Google APIs! What we would love to see is more of Google in the Google APIs rather than it seeming the work of separate teams or project groups.  Key to this is more proactive marketing and communication in terms of transitioning from the status quo, and more education on how various bits and pieces tie in together. It would be better to see  a more broad range of case studies, customer success stories ,demos and examples in the Gallery section of each API. It can help us with more confidence in recommending this to enterprise customers than just pointing to a series of blandly written developer.google.com webpages. Perhaps Mr Page can borrow a webpage from Metamarkets.

An additional feature request would be PostgreSQL  in the cloud , and not just MySQL in the cloud as a service. You can have a look at CitusData and see how that is awesome too

RDBMS in the cloud - just another API call away- though much work needs to be done here.

Ajay Ohri is the author of R for Business Analytics and likes to write on Enterprise ,Cloud and Statistical APIs with an emphasis on interviews. Follow Ajay on Google+ and connect on LinkedIn

Comments

Comments(4)