In the last two decades, the continuous increase of computational power has produced an
overwhelming flow of data which has called for a paradigm shift in the computing …

MapReduce is a programming system for distributed processing of large-scale data in an
efficient and fault tolerant manner on a private, public, or hybrid cloud. MapReduce is …

While high-level data parallel frameworks, like MapReduce, simplify the design and
implementation of large-scale data processing systems, they do not naturally or efficiently …

We offer a new metric for big data platforms, COST, or the Configuration that Outperforms a
Single Thread. The COST of a given platform for a given problem is the hardware …

Over half a century old and showing no signs of aging, k-means remains one of the most
popular data processing algorithms. As is well-known, a proper initialization of k-means is …

How can one summarize a massive data set" on the fly", ie, without even having seen it in its
entirety? In this paper, we address the problem of extracting representative elements from a …

The algorithms community has been modeling the underlying key features and constraints of
massively parallel frameworks and using these models to discover new algorithmic …

Many large-scale machine learning problems (such as clustering, non-parametric learning,
kernel machines, etc.) require selecting, out of a massive data set, a manageable …

We initiate the study of graph sketching, ie, algorithms that use a limited number of linear
measurements of a graph to determine the properties of the graph. While a graph on n …

Greedy algorithms are practitioners' best friends—they are intuitive, are simple to implement,
and often lead to very good solutions. However, implementing greedy algorithms in a …