ignite-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Dave Harvey <dhar...@jobcase.com>
Subject Ignite/yardstick benchmarking model question
Date Mon, 11 Sep 2017 21:33:55 GMT
I've started looking at the standard benchmarks in
apache-ignite-fabric-2.1.0-bin/benchmarks, and I'm trying to understand the
relationship between the "Benchmark server" and "Ignite server".   Does
starting the "Benchmark server" start the "ignite server" or are those
different?   Staring at the code has not yet made this clear.

I have setup a cluster with 1 Ignite client host, and 4 Ignite server hosts.   
The servers are running, and communicating.  The client host does not have
ssh access to the server hosts.    I want to run a benchmark where the
benchmark code runs on the client host only.   I do not want any yardstick
code on the Ignite servers.   I do not want to be starting and stopping the
Ignite server nodes.

It's unclear to me whether I can configure the existing higher Ignite
benchmarks to do what I would like, or I have to do something with the
piece-parts, such as bin/benchmark-drivers-start.sh.     I'm trying to get
at the performance/behavior of the key/value interface being called on the
client node, when all of the storage is on the server nodes.

Sent from: http://apache-ignite-users.70518.x6.nabble.com/

View raw message