Transformer: A New Paradigm for Building Data-Parallel Programming Models

19 August 2010

journal article
Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Micro

Vol. 30 (4) , 55-64
https://doi.org/10.1109/mm.2010.75

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

This publication has 9 references indexed in Scilit:

Pregel
Published by Association for Computing Machinery (ACM) ,2009
The Datacenter as a Computer: An Introduction to the Design of Warehouse-Scale Machines
Synthesis Lectures on Computer Architecture, 2009
Pig latin
Published by Association for Computing Machinery (ACM) ,2008
All-pairs: An abstraction for data-intensive cloud computing
2008 IEEE International Symposium on Parallel and Distributed Processing, 2008
Technical perspective
Communications of the ACM, 2008
Dryad
Published by Association for Computing Machinery (ACM) ,2007
Interpreting the Data: Parallel Analysis with Sawzall
Scientific Programming, 2005
Advances in dataflow programming languages
ACM Computing Surveys, 2004
Actors
Published by MIT Press ,1986