Most big data strategies focus on storage and analytics, with the collection and distribution of all that information an afterthought. Using real-world examples of distributed big data systems in capital markets trading systems, mobile carrier customer service systems, and M2M sensor networks, this talk will discuss the value and challenges of efficiently moving information at big data scale. From the aggregation of distributed data for loading into Hadoop clusters to synchronizing terabyte-scale in-memory data grids across geographies, Ken will discuss techniques and tools that can increase the rate and efficiency of data flows within big data architectures.