Most data analysis and data mining today takes place using client server technology applied to local data, despite the fact that during the past several years a number of technologies and frameworks have emerged for working with remote and distributed data. Although bandwidth is becoming a commodity and the amount of digital data continues to grow exponentially, it is still remarkably hard to access, explore, analyze and mine remote and distributed data. In this talk, we discuss some of the underlying reasons and survey some of the competing architectures and platforms for distributed data mining, including:
We end the talk by describing some current research problems facing the field.