Solving Problems with Graphs by Marko A. Rodriguez.
Marko covers solving problems with graphs in general and then gives an overview of Titan (a distributed graph database), Faunus (graph analytic engine) and Fulgora (graph processor).
My only misgiving about graphs is that we know very little of the world’s data is stored in graph format. And that is unlikely to change in the foreseeable future. ETL will suffice convert some data to obtain the advantages of graph processing, but what of data that isn’t converted?
Unlike the W3C, I have a high degree of confidence that the world is not going to adapt itself to any one solution or even a range of solutions.
The majority of data (from a current perspective), will be in “legacy” formats, the next largest portion in the successful formats just prior to the latest one, and the smallest portion, the latest proposed new format.
Big data should address the “not my format” problem in addition to running after large amounts of sensor data.