command line interface with default parameters
command line interface with default parameters
if the number of remaining edges are lower than this value, pacc run a single machine algorithm (LocalCC)
the number of partitions
input file path
output file path
CC-Computation operation.
CC-Computation operation. This operation conducts LocalCC in each partition.
input RDD
the number of partitions
final output RDD containing connected components
PA-Large-Star Operation.
PA-Large-Star Operation. For each node n, this operation links each large neighbor v to the minimum node mcu(p(v)) in the same partition p(v) that contains the neighbor v.
the input rdd
the number of partitions
current round number
the temporary path to save the intermediate results
(RDD for next round input, # changed edges, # of 'out' edges, # filtered 'cc' edges, # filtered 'in' edges)
submit the spark job.
submit the spark job.
input file path.
if the number of remaining edges are lower than this value, pacc run a single machine algorithm (LocalCC).
the number of partitions
spark context.
an RDD containing connected components
run by command line interface
run by command line interface
PA-Small-Star Operation.
PA-Small-Star Operation. For each node n, this operation links each small neighbor v to the minimum node mcu(p(v)) in the same partition p(v) that contains the neighbor v.
the input rdd
the number of partitions
current round number
the temporary path to save the intermediate results
(RDD for next round input, # changed edges, # of 'out' edges, # filtered 'in' edges)