What is NoSQL? Could it be database systems that only use pairs?

So far as I understand MemCache is just one of such database systems, am I right?

The other popular NoSQL databases exist and where exactly could they be helpful?

Thanks, Boda Cydo.

From wikipedia:

NoSQL is definitely an umbrella term for any loosely defined class of non-relational data stores that break having a lengthy good reputation for relational databases and Acidity guarantees. Data stores that come under this term might not require fixed table schemas, in most cases avoid join procedures. The word was initially popularised at the begining of 2009.

The motivation for this kind of architecture was high scalability, to aid sites for example Facebook, advertising.com, etc...

To rapidly get a grip on NoSQL systems, see my Visual Help guide to NoSQL Systems. Basically, NoSQL systems sacrifice either consistency or availability in support of ability to tolerate network partitions.

I made use of something known as the Raima Data Manager greater than a dozen years back, that qualifies as NoSQL. It calls itself a "Set Oriented Database" It is not according to tables, and there's no query "language", just an C API for requesting subsets.

It's fast and simpler to utilize in C/C++ and SQL, there is no accumulating strings to pass through to some query interpreter and also the data returns being an enumerable object instead of being an array. variable sized records are common and do not waste space. I never saw the origin code, but there have been some shows the interface that internally, the code used pointers a great deal.

I am unsure the product I made use of is even offered any longer, but the organization remains to be.

Jon Meredith lately gave an exhibition towards the Front Range PHP Customers Group about NoSQL databases that you find helpful: link

MongoDB looks interesting, SourceForge has become utilizing it.

I took in to some podcast having a team member. The concept with NoSQL is not a lot to exchange SQL because it is use a solution for issues that aren't solved well with traditional RDBMS. As pointed out elsewhere, they're faster and scale better at the expense of reliability and atomicity (different methods to different levels). You wouldn't like to make use of one for any economic climate, but a document based system works great.