Class Summary |
HashIdMapper |
This mapper class maps an (vid, vdata) pair into (lvid, (vid, vdata)) pair. |
HashIdMR |
This MapReduce class maps a list of unique vertex into 2 parts of output: A
dictionary from rawId to newId, and a new vertex data file using newId. |
HashIdReducer |
Reducer class reduce the (baseid, List<(vid, vdata)>) into a dictionary
(lvid, vid), and a new vertex data file (lvid, vdata), where lvid = baseid +
vid.index * splitsize; Because the splitsize is fixed for all mapper using
option mapred.line.input.format.linespermap in the JobConf ,
this guarantee all vids are mapped into [0, ..., |V|-1]. |
SortDictMapper |
Mapper function for SortDictMR job |
SortDictMR |
This MapReduce class partitions the dictionary output of HashIdMR based on
the hash of the rawId, the key. |
SortDictReducer |
Reducer function for SortDictMR job |
SortEdgeMR |
This class partition the edge list input by the hash of the source vertex. |
SortEdgeMR.SortEdgeMapper |
This mapper class maps each edge into (h(edge.source), edge). |
SortEdgeMR.SortEdgeReducer |
This reducer class takes the input (hashval, edge) from mapper and outputs
edge directly. |
TransEdgeMapper<VidType extends WritableComparable<VidType>> |
This mapper class maps each edge as (u,v,data) into (h(v), (D(u), v, data))
where D is the dictionary that contains entry u. |
TransEdgeMR |
This MapReduce class translate the rawIds in the edge list into normalized
newIds using the partitioned edgelist output from SortEdgeMR and partitioned
dictionary output from SortDictMR. |
TransEdgeReducer<VidType extends WritableComparable<VidType>> |
This reducer class takes from mapper the input (h(v), [(D(u_i),v_i,
data)...]) and output (D(u_i), D(v_i), data). |