Title :
Transformer: A New Paradigm for Building Data-Parallel Programming Models
Author :
Wang, Peng ; Meng, Dan ; Han, Jizhong ; Zhan, Jianfeng ; Tu, Bibo ; Shi, Xiaofeng ; Wan, Le
Author_Institution :
Inst. of Comput. Technol., Chinese Acad. of Sci., Beijing, China
Abstract :
Cloud computing drives the design and development of diverse programming models for massive data processing. the transformer programming framework aims to facilitate the building of diverse data-parallel programming models. transformer has two layers: a common runtime system and a model-specific system. using transformer, the authors show how to implement three programming models: dryad-like data flow, MapReduce, and All-Pairs.
Keywords :
parallel programming; All-Pairs processing; MapReduce processing; cloud computing; common runtime system; data-parallel programming models; dryad-like data flow processing; model-specific system; transformer programming framework; Computational modeling; Data models; Libraries; Load modeling; Programming; Receivers; Runtime; All-Pairs; MapReduce; actor model; cloud computing; data flow; data intensive computing; programming model;
Journal_Title :
Micro, IEEE
