Data processing technologies used in servers are the critical foundations for the cloud computing systems.This paper introduces the large scale data processing technologies used in the current cloud computing implementations.Such technologies are quite important for the data collections as well as processing.There are three key goals for processing the large scale data i.e.reliability
scalability and easy to program.Based on the Map Reduce programming paradigm from Google and Dryad model from Microsoft
how to achieve the goals in practice are discussed in detail.
关键词
Keywords
references
Interpreting the data: Parallel analysis with Sawzall [J] . Carlos A. Varela,Paolo Ciancarini,Kenjiro Taura,Rob Pike,Sean Dorward,Robert Griesemer,Sean Quinlan. Scientific Programming . 2005 (4)