Minisymposium Presentation

Computing platforms are evolving rapidly along many dimensions: processors, specialization, disaggregation, acceleration, smart memory and storage, etc. Many of these developments are being driven by data science but very few existing data processing platforms make use of modern hardware, which is crucial to be able to process large amounts of data in an efficient manner. One reason for the gap is the deluge of possible configurations and deployment options, most of them too new to have a precise idea of their performance implications and lacking proper support in the form of tools and platforms that can manage the underlying diversity. This growing heterogeneity opens up many opportunities but also raises significant challenges. In the talk I will describe our efforts to explore the possibilities that modern hardware opens for data management and discuss a system we are building for data processing on heterogeneous computing platforms that has as its main goal to effectively cope with the great variety of emerging hardware.