I'm trying to use the Cassandra stress tool but I don't know the unit for the size distribution for a blob column.
The blog post doesn't really help in this regard either.
So the question is when I specify a columnspec
on a blob column like this:
columnspec:
- name: data
size: gaussian(10000..250000, 2)
What units are the min/max referring to? Is this bits, bytes, MB, GB?
If I understand this code correctly (and assuming this is where I should be looking), particularly the constructor:
public HexBytes(String name, GeneratorConfig config)
{
super(BytesType.instance, config, name, ByteBuffer.class);
bytes = new byte[(int) sizeDistribution.maxValue()];
}
It is initializing a byte[]
with the distribution's max value.
In that case the size would correspond to bytes.