Re: Query faster 10 times when stop on data node

[Mikael Ronström]

Hi,
When you run with 2 data nodes, then half of the data in the query is
fetched from the remote node. When only the data node on the same server
is alive, then all data is fetched from this data node.

Why exactly this gives a 30x improvement is though hard to say, to answer
this requires better understanding of both the query itself as well as the
HW environment you are running in.

Rgrds Mikael Ronström