5. Graph Processing Paradigms
As discussed in Chapter 1, “Introduction: Why Look Beyond Hadoop Map-Reduce?,” giant 4 (graph processing) requires specialized paradigms for processing. One such paradigm is the Bulk Synchronous Parallel (BSP), proposed by Leslie Valiant (1990). There are several realizations of the BSP in the literature, with Google’s Pregel being the precursor to some of the other tools. Apache Giraph is the open source equivalent of Pregel, and Apache Hama is another similar work. We will look at a few graph-processing tools starting with Pregel. We start with a discussion of what the need is for graph-processing paradigms.
Facebook has recently opened up a search feature (officially known as a graph search) that allows people ...
Get Big Data Analytics Beyond Hadoop: Real-Time Applications with Storm, Spark, and More Hadoop Alternatives now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
