hadoop behind the scenes -
What can anyone explain in terms of the ideas behind the software? Does it make this so popular and / or powerful?
Hadop is a programming environment in which large scale computation runs parallel to a large cluster of machines, Is flexible for the loss of, Scalable to enable computation to speed up the machines to add and report computing status. Hadop is popular because it is a strong open source environment and because many users, including large users, such as Yahoo !, Microsoft and Facebook, use it for large data-crunching projects, because it is less powerful than map / algorithm , Which breaks down a calculation in the sequence of two simple tasks:
- Map - Take a list of items and each one has the same simple opera Find them Take the text of a web page, for example, token it and replace each token with string: 1
- Reduce - Take a list of items and submit it using a accumulation operator. For example, take a list of 1: Calculate a list of forms and output: nt, where nt is visible several times in the original list.
Proper usage decomposition (which makes the programmer) and work distribution and monitoring (which Hadoop does) gives you a fast scalable computation; In our example - a word count calculation. You can sequence tens of maps and reduce and can get the implementation of sophisticated algorithms. This is a very high level view. Now read more about it in detail.
Comments
Post a Comment