Key points are not available for this paper at this time.
We present Sketch, a library and a distributed runtime for building interactive tools for exploring large datasets, distributed across multiple machines. We have built several applications using Sketch; here we describe a billionrow spreadsheet, and a distributed-systems performance analyzer. Sketch applications allow interactive and responsive exploration of complex distributed datasets, scaling effectively to use large computational resources.
Budiu et al. (Fri,) studied this question.