[Gluster-users] Cassandra Stress Testing

Mohit Anchlia mohitanchlia at gmail.com
Mon Mar 21 19:49:22 UTC 2011


Thanks!

I ran mvn compile install and produced
/drives/c/proj/zznate-cassandra-stress-90ff926/target/cassandra-stress-0.7_25.jar.

When I try to run:
C:\proj\zznate-cassandra-stress-90ff926>java -jar target\cassandra-stress-0.7_25
.jar
Failed to load Main-Class manifest attribute from
target\cassandra-stress-0.7_25.jar

Am I doing something wrong?

On Mon, Mar 21, 2011 at 12:24 PM, Nate McCall <nate at datastax.com> wrote:
> RF is at the config level of the cluster. CL is not yet configurable
> via the CLI. I have something close to that locally which I can add
> shortly here though.
>
>
>
> On Mon, Mar 21, 2011 at 2:03 PM, Mohit Anchlia <mohitanchlia at gmail.com> wrote:
>> Is RF and CL also configurable or do I need to make code changes for that?
>>
>> On Mon, Mar 21, 2011 at 11:43 AM, Mohit Anchlia <mohitanchlia at gmail.com> wrote:
>>> I'll take a look at the code and see. But first I will to setup and compile :)
>>>
>>> On Mon, Mar 21, 2011 at 11:32 AM, Nate McCall <nate at datastax.com> wrote:
>>>> Never tried with large blobs. The code generates a random string of
>>>> 'w' length, (it's all binary on the wire at the end of the day, string
>>>> vs. bytes should not matter too much), so give it a try and let me
>>>> know how it goes :-)
>>>>
>>>> -N
>>>>
>>>> 2011/3/21 Patricio Echagüe <patricioe at gmail.com>:
>>>>>  options.addOption("w","colwidth",true,"The widht of the column in bytes.
>>>>> Default is 16");
>>>>> -w <column_size> is the column size.  But I think for better blob
>>>>> manipulation you will have to extend the framework a bit.
>>>>>
>>>>> Nate, any light on this?
>>>>>
>>>>> On Mon, Mar 21, 2011 at 11:18 AM, Mohit Anchlia <mohitanchlia at gmail.com>
>>>>> wrote:
>>>>>>
>>>>>> Thanks!
>>>>>>
>>>>>> Can I also give the size of the blobs I want to test with? Is
>>>>>> stress-schema configurable to what I need to test? Can I give min, max
>>>>>> and avg sizes of blobs in the test?
>>>>>>
>>>>>> 2011/3/21 Patricio Echagüe <patricioe at gmail.com>:
>>>>>> > https://github.com/zznate/cassandra-stress
>>>>>> >
>>>>>> > On Mon, Mar 21, 2011 at 10:00 AM, Mohit Anchlia <mohitanchlia at gmail.com>
>>>>>> > wrote:
>>>>>> >>
>>>>>> >> One of the use cases I need to performance test is the write and read
>>>>>> >> of blobs (avg 64K, max 30 MB). Is there a hector tool I can use or
>>>>>> >> should I just write simple hector clients of my own?
>>>>>> >
>>>>>> >
>>>>>
>>>>>
>>>>
>>>
>>
>



More information about the Gluster-users mailing list