A 101 Guide to Heterogeneous, Accelerated, Data Centric Computing Architectures

Abstract

Over the last few years several terminologies have become popularized in the world of computing. Namely, the words Heterogeneous, Accelerated and Data Centric are far more frequently used than ever before. These terms are generally understood but the underlying reasons for this shift in the lexicon has not really been well communicated. This presentation aims to debunk the complexity of the language with a 101 guide and explanation of the disadvantages of traditional Von Neumann, CPU Centric Architectures and their 25 year history of “patching up” and “working around” the well understood Von Neuman Memory Wall Syndrome with increasingly complex caching hierarchies and structures. The presentation will close with highlighting how IBM OpenPower Foundation members are all feverishly collaborating to make this seismic shift in the computing industry happen as smoothly and quickly as possible to bring far more computational horse power for far less energy & cost become a true reality.