Container class to store an id and value

Map reduce job to compress a graph in such a way that each community is represented by a single node. <p/> input format: see LouvainVertexOutputForma

Abstract class that simplifies the setup of our EdgeInputFormat subclasses. <p/> All DGA analytics require data to be specified in roughly the same wa

<p>The EdgeTDTOutputFormat outputs the edges that make up our graph.<p/> <p>This output format will allow users to write out the source vertex ID, the

Simple implementation that offloads work of parsing to the RawEdge class and the work of casting our edgeValue as a Writable class of choice to the im

A Edge Writer that writes each edge into a file on HDFS.

Implementation of the DGAAbstractEdgeInputFormat for LongWritables as the edgeValue Will throw an IOException if our edgeValue can't be parsed to Long

Implementation of the DGAAbstractEdgeInputFormat for Text as the edgeValue Does no verification of the edgeValue and simply returns the third value as

Calculates Shortest paths, and accumulates pair dependency information. <p/> Use with SBMasterCompute to find an approximated set of vertices with hig

Coordinates Global state for the SBVertex to calculate shortest paths, accumulates pair dependency information, and monitor high betweenness set stabi

Writes the approximated betweenness value for each vertex

Maintains a list of the top N items(an item is defined as an int id, and double value), ranked by a double value. Designed for use with the giraph Agg

Aggregator to accumulate a list of the top N vertices, ranked by approximated betweenness value.

Leaf Compression is an analytic used to compress a graph; nodes on the periphery of the graph that do not show an extensive network of connections fro

Performs the BSP portion of the distributed louvain algorithm. <p/> The computation is completed as a series of repeated steps, movement is restricted

Master compute class. performs a compute function before each super step. Performs 4 functions. <p/> 1. prints to its standard out the number of node

messages sent between vertices.

The state of a vertex.

Reads in a graph from text file in hdfs. Required format in is a tab delimited file with 3 columns: <code>id<tab>internal weight<tab>edge list the ed

Outputs the graph as text in hdfs: <p/> Format is a tab seperated file with id community id internal weight community edge list <p/> the edge list is

Writable class to represent community information for compressing a graph by its communities.

Represents a source vertex's dependency on a specific target vertex. <p/> For more information about pair dependencies and betweenness centrality see

Message that can be passed between vertices. Used for both shortest path computation and pair dependency accumulation. Some fields are only used in

A simple vertex writer that writes the Vertex and it's HighBetweenness Value.

Keeps Shortest path data for a single source vertex to a single target vertex. <p/> Maintains Shortest path, predecessors, and number of shortest path

A simple vertex writer that writes the Vertex and it's HighBetweenness Value.

Global States that direct certain computation. <ul> <li>START: Chooses the initial batch size.</li> <li>SHORTEST_PATH_START: Starts to calculate the s

A Simple Edge Writer that writes each edge into a file on HDFS.

A Simple Edge Writer that writes each edge into a file on HDFS.

Stores all required data for a vertex <ul> <li>Map of Sources to ShortestPathLists (for shortest path phase)</li> <li>Map of Sources to PartialDepende

WeaklyConnectedComponents is the concept of finding how many uniquely connected nodes there are in a specific data set.