Hbase Heap Size Calculator

hbase_heap.py

def get_regionserver_heap_size(
    storage_capacity_in_gb,
    region_max_filesize=10737418240,
    memstore_flush_size=134217728,
    replication_factor=3,
    memstore_heap_fraction=0.4
):
    """
    Calculates heap size required based on storage requirements.
    
    Source: http://hadoop-hbase.blogspot.com.au/2013/01/hbase-region-server-memory-sizing.html
    
    Args:
        storage_capacity_in_gb (int): Total disk capacity per box.
        region_max_filesize (int): Region size (hbase.hregion.max.filesize)
        memstore_flush_size (int): Memstore flush size (hbase.hregion.memstore.flush.size)
        replication_factor (int): HDFS replication factor (dfs.replication)
        memstore_heap_fraction (int): Memstore heap fraction (hbase.regionserver.global.memstore.size)
    """
    one_gb_in_bytes = 1073741824.0
    storage_capacity_in_bytes = storage_capacity_in_gb * one_gb_in_bytes
    number_of_regions = (1.0 * storage_capacity_in_bytes / replication_factor) / region_max_filesize
    return number_of_regions * memstore_flush_size / memstore_heap_fraction / one_gb_in_bytes