Database in the Cloud

I go away for a week to Amsterdam (last week) while the annual SFDC (SalesForce.com) user event called Dreamforce took place in San Francisco. A significant announcement there was a new product called Database.com – simply a database in the cloud. Immediately the press pounced on this as much more than what is real, such as  – this is a threat to Oracle’s dominating database product, etc.

The features of this product have been under use inside Force.com platform, meant for developers to build any application that needs to be hosted as a SaaS. In other words, Database.com is a plug-and-play storage infrastructure for any developer. It has standard features like user management, row-level security, triggers and stored procedures, a query language and search functions. As this will reside in the cloud, it has autonomic tuning, backup, replication and upgrades. It supports open APIs such as REST, SOAP, oAuth, etc. Developers can write apps. in any language like Java, C#, Ruby, or PHP.  The runtime could operate on any of these – Force.com, VMForce, Amazon EC2, Google App. Engine, Microsoft Azure or Heroku.

Just like Software as a Service, or Platform as a Service, there is need for Database as a Service (DaaS).  Database.com is an attempt by SFDC to fulfill this need. It is more like Microsoft’s SQL Azure than Oracle’s DBMS. Pricing of Database.com is again made attractive for wider adoption – free for up to 3 users with 100K records and 50K transactions per month. Then the price goes in $10/month increment at certain thresholds.

I am involved with another start-up called ScaleDB that provides real scale-out for high volume workloads in the cloud. This product uses MySQL open source and adds very clever clustering technology. Clearly there is need for a Database in the Cloud. How robust is Database.com for fast-scaling application workloads? Let’s wait and see. Today it’s all marketing stuff.