The Revolution in Database Architecture, by Jim Gray

The Revolution in Database Architecture, by Jim Gray, describes the path that Gray thought that the evolution of the Database Architecture would take after 2004. He considers that databases had been stagnated for several years and that, beginning in 2004, the development of several technologies would pave the way into a revolution in the database world.

The 14 technologies mentioned as the key for the revolution are:

1. The ability to execute Object Oriented code inside the RDBMS.

2. Databases becoming a web service. E.g. a database exposes an interface in the web that allows direct execution of store procedures.

3. Queues becoming an integral part of the Database, therefore allowing Applications to be loosely connected via Queued messages.

4. Cubes and OLAP, providing a simple way to aggregate data.

5. Data mining, to extract patterns and key information from databases.

6. Column stores, providing a more compact way of storing data (for records with many columns, most of them unused)

7. Databases having better support for unstructured data (i.e. Text, Spatial and Temporal data).

8. Semi structured data, for dealing with data that doesn’t “fit” in the relational model.

9. Stream processing, for comparing incoming data about an object against the history of the object.

10. Publish/Subscribe pattern, in which many users subscribe to a data warehouse which publishes the data they need.

11. Query optimizers becoming dynamic i.e. “changing plans as the system load and data statistics change”

12. Database in main memory to circumvent the growing ratio of storage capacity/latency of hard drives.

13. Taking advantage of the CPU’s on most devices to turn them into a small DBMS. This would permit everything to share data w...

... middle of paper ...

...deserved.

That being said, Gray’s style of writing gets the point across easily. He manages to give the reader a rough idea of how 14 technologies work, the problems they come to solve and how they will revolutionize the database world. Such a succinct writing is commendable.

Concluding Remarks

Gray’s article is interesting. Some of his predictions of the future were off, and he made them seem more transcendent than they actually are. However, predicting the future is no easy task, and Gray did a remarkably good job out of it.

Errors such as fallacies in the middle of a scholarly article are unacceptable. So is not providing references and the source of information used for building the article. PhD or not, Gray shouldn’t have taken this luxury. He should set an example on both of these accounts.

Works Cited

The Revolution in Database Architecture