hadoop - HDFS: Actual space of all disks on cluster vs usable HDFS size -
how can calculate available size of hdfs cluster based on total size of disks in cluster?
e.g. if cluster 10 machines, each 1tb of storage, hadoop fs -df
report?
more specifically, need store 5 tb of data in hdfs cluster. how total disk space cluster need?
that depends on how set hdfs replication factor. default (and recommended) 3.
you can set hdfs keep non-dfs space so, substract if need be.
a rough calculation file size * 3 = total storage needed
.
Comments
Post a Comment