More than a decade ago, the pioneering computer scientist Jim Gray envisioned a fourth paradigm ofscientific discovery that uses computers to solve data-intensive scientific problems. Today, scientificdiscovery, whether observational, in-silico or experimental, requires sifting through and analyzingcomplex, large datasets. For example, plasma simulation simulates billions of particles in a single run,but analyzing the results requires sifting through a single frame (a multi-dimensional array) that is morethan 50 TB big—and that is for only one timestep of a much longer simulation. Similarly, modernobservation instruments also produce large datasets: a two-photon imaging of a mouse brain yields upto 100 GB of spatiotemporal data per hour and electrocorticography (ECoG) recordings yield 280 GB perhour.
展开▼
