Hi, I¹m a Solution Architect at DataStax. I¹ve written a Cassandra plugin for Jmeter with a GUI config element, sampler, and pre- and post- elements. It¹s fairly similar to the JDBC sampler in terms of it¹s functionality. In addition, I have a detailed summarizer that returns the statistics for each sampler in non-gui mode, so long as the sampler names start with <num>:, as it uses an array to collect the data. The delimiter and the array size are all configurable in the jmeter.properties file.
The output looks like this: 0: Write trades 20034 in 30s = 668.0/s Avg: 2 Min: 0 Max: 56 Err: 0 (0.00%) 1: Write trades_by_tickerday 20032 in 30s = 667.9/s Avg: 2 Min: 0 Max: 60 Err: 0 (0.00%) 2: Write trades_by_datehour 20031 in 30s = 667.9/s Avg: 2 Min: 0 Max: 52 Err: 0 (0.00%) 3: Trade by id 19333 in 30s = 644.4/s Avg: 4 Min: 0 Max: 59 Err: 0 (0.00%) 4: 10 Minute Range By Ticker First 1 19212 in 30s = 640.6/s Avg: 4 Min: 0 Max: 59 Err: 0 (0.00%) 5: 1 Minute Range By Hour First 1 19324 in 30s = 644.0/s Avg: 6 Min: 0 Max: 69 Err: 0 (0.00%) summary: + 117966 in 30s = 3931.5/s Avg: 3 Min: 0 Max: 69 Err: 0 (0.00%) Active: 18 Started: 18 Finished: 0 summary: = 139266 in 42s = 3345.1/s Avg: 3 Min: 0 Max: 84 Err: 0 (0.00%) I have everything in a private github repo, but I checked in the vanilla 2.13 stuff so I can easily extract the changes. In that repo, I have the sample changes, and some packaging changing to include the cassandra plugin as well as the groovy jar. The plugin itself is currently in its own public repo, and is built into a nice zip package which contains the Jar file and its dependencies. How do I go about properly integrating these changes into JMeter? Steve.
