Yale Researchers Prove That ACID Is Scalable

An anonymous reader writes "The has been a lot of buzz in the industry lately about NoSQL databases helping Twitter, Amazon, and Digg scale their transactional workloads. But there has been some recent pushback from database luminaries such as Michael Stonebraker. Now, a couple of researchers at Yale University claim that NoSQL is no longer necessary now that they have scaled traditional ACID compliant database systems."

Comments

Re:digg does not need to worry anymore

[cawpin]

I cannot even put into words how ignorant-of-the-situation that statement is.

Re:Pfah.

[Requiem18th]

You are missing the point.

For aggregation, Relational is king, an RDBMS is *made* to answer questions such as "Find me all people who own a 1996-2010 Year Toyota Prius.", it is a great tool for analysts, admins managers and marketroids.

What an RDBMS is not made for that a Non-RDBMS is, is answering a million simultaneous questions of the form, "Find me the list of cars of user $username as well as his/her name, gender, email, latest picture, favorite color, friends list and whether he said he prefers cats or dogs in our last survey if he did took such survey." as well as saving back any changes the user makes.

An ACID compliant RDBMS can't even get read access to the user, car, friend, picture and pet_survey_answer table set as long as any of the million users of the system is making a change to his data, even if the application only locks one table at a time for write access, let alone the problem of a million users trying to gain write access to the same table at the same time.

All of this because the application's persistence model is optimized for offline analysis rather than, well providing persistence for the application.

So it's not like Non-RDBMS only exist because of ignorance and incompetency.