[
https://issues.apache.org/jira/browse/CASSANDRA-8597?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14273835#comment-14273835
]
Ariel Weisberg commented on CASSANDRA-8597:
-------------------------------------------
My use case was simple so I didn't run into capability issues. My comments are
almost all about documentation.
The embedded help is very good although concepts can be hard. It's the higher
level docs I had trouble with. The first was that I didn't find the right
stress docs because I got 2.0 when I had to go specifically look for 2.1 docs.
Embedded help providing a doc URL would be nice.
Flag formatting is tricky both due to how positional they are, but also due to
the escaping required in the shell for parentheses. When you are working
through it the first time there are small things like that breaking your flow.
There was also the fact that distributions don't accept spaces between
parameters.
The last bit is the lack of recipe style documentation. Most flags are part of
some recipe so it's helpful to see them in action.
There are two dimensions data distribution and access distribution and knowing
all the knobs for describing data distribution (# partition, # cells/rows,
#size of cells) and access distribution (across partitions, across cells within
partitions, random vs sequential strides, # of rows/cells to select) would be
helpful. I think some times these parameters are linked as well.
It's also worth mentioning in the distribution section that you can get the
tool to print a summary of distributions so you know what your parameters are
doing.
> Stress: make simple things simple
> ---------------------------------
>
> Key: CASSANDRA-8597
> URL: https://issues.apache.org/jira/browse/CASSANDRA-8597
> Project: Cassandra
> Issue Type: Improvement
> Components: Tools
> Reporter: Jonathan Ellis
> Assignee: T Jake Luciani
> Fix For: 2.1.3
>
>
> Some of the trouble people have with stress is a documentation problem, but
> some is functional.
> Comments from [~iamaleksey]:
> # 3 clustering columns, make a million cells in a single partition, should be
> simple, but it's not. have to tweak 'clustering' on the three columns just
> right to make stress work at all. w/ some values it'd just gets stuck forever
> computing batches
> # for others, it generates huge, megabyte-size batches, utterly disrespecting
> 'select' clause in 'insert'
> # I want a sequential generator too, to be able to predict deterministic
> result sets. uniform() only gets you so far
> # impossible to simulate a time series workload
> /cc [~jshook] [~aweisberg] [~benedict]
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
