Runs a Bagel program with no org.apache.spark.bagel.Aggregator, the default org.
Runs a Bagel program with no org.apache.spark.bagel.Aggregator, the default org.apache.spark.HashPartitioner and org.apache.spark.bagel.DefaultCombiner
Runs a Bagel program with no org.apache.spark.bagel.Aggregator, default org.
Runs a Bagel program with no org.apache.spark.bagel.Aggregator, default org.apache.spark.HashPartitioner, org.apache.spark.bagel.DefaultCombiner and the default storage level
Runs a Bagel program with no org.apache.spark.bagel.Aggregator and the default org.
Runs a Bagel program with no org.apache.spark.bagel.Aggregator and the default org.apache.spark.HashPartitioner
Runs a Bagel program with no org.apache.spark.bagel.Aggregator, default org.
Runs a Bagel program with no org.apache.spark.bagel.Aggregator, default org.apache.spark.HashPartitioner and default storage level
Runs a Bagel program with no org.apache.spark.bagel.Aggregator
Runs a Bagel program with no org.apache.spark.bagel.Aggregator and the default storage level
Runs a Bagel program.
Runs a Bagel program.
key
vertex type
message type
combiner
aggregator
org.apache.spark.SparkContext to use for the program.
vertices of the graph represented as an RDD of (Key, Vertex) pairs. Often the Key will be the vertex id.
initial set of messages represented as an RDD of (Key, Message) pairs. Often this will be an empty array, i.e. sc.parallelize(Array[K, Message]()).
org.apache.spark.bagel.Combiner combines multiple individual messages to a given vertex into one message before sending (which often involves network I/O).
org.apache.spark.bagel.Aggregator performs a reduce across all vertices after each superstep and provides the result to each vertex in the next superstep.
org.apache.spark.Partitioner partitions values by key
number of partitions across which to split the graph. Default is the default parallelism of the SparkContext
org.apache.spark.storage.StorageLevel to use for caching of intermediate RDDs in each superstep. Defaults to caching in memory.
function that takes a Vertex, optional set of (possibly combined) messages to the Vertex, optional Aggregator and the current superstep, and returns a set of (Vertex, outgoing Messages) pairs
an RDD of (K, V) pairs representing the graph after completion of the program