CloudBurst Wiki
Brought to you by:
mcschatz
Usage: CloudBurst refpath qrypath outpath readlen k allowdifferences filteralignments #mappers #reduces #fmappers #freducers blocksize redundancy
refpath: path in hdfs to the reference file
qrypath: path in hdfs to the query file
outpath: path to a directory to store the results (old results are automatically deleted)
readlen: minimum length of the reads
k: number of mismatches / differences to allow (higher number requires more time)
allowdifferences: 0: mismatches only, 1: indels as well
filteralignments: 0: all alignments, 1: only report unambiguous best alignment (results identical to RMAP)
#mappers: number of mappers to use. suggested: #processor-cores * 10
#reduces: number of reducers to use. suggested: #processor-cores * 2
#fmappers: number of mappers for filtration alg. suggested: #processor-cores
#freducers: number of reducers for filtration alg. suggested: #processor-cores
blocksize: number of qry and ref tuples to consider at a time in the reduce phase. suggested: 128
redundancy: number of copies of low complexity seeds to use. suggested: # processor cores