gistfile1.txt

12:27 <nginr> Hi .. Need a suggestion. I want to test the behaviour of my 
cluster as the data in the cluster keeps on increasing. Upto 1 TB. 
I want to test the behaviour of Read, Write and Search. Is it possible to do with curl ?? 

12:28 <chids> nginr: Possible to do what with curl - execute read, 
write and search requests? 

12:28 <aphyr> PBC or erlang might be faster, but I've used the HTTP interface 
for testing in our ruby app 

12:31 <nginr> chids: Yes . read, write and identify the time for read, write. 
If I want to test for data upto TBs, what should be my key and value ? 

12:32 <aphyr> nginr: What does your intended data look like? Probably makes 
sense to simulate something with similar characteristics 

12:32 <roidrage> nginr: do you want a single piece of data to be 1tb in size, 
or are you just going to fill up the cluster with smaller chunks of data over time? 

12:32 <nginr> roidrage: I m going to fill smaller chunks of data for a long time .. 

12:33 <roidrage> nginr: i'd suggest you should check out basho_bench: 
https://github.com/basho/basho_bench 

12:34 <roidrage> and in general you could certainly use curl, but you should 
also consider using a protobuf interface. depends on your use case in general, 
but it's the better performing interface 

12:35 <nginr> roidrage: But the application we are creating uses only http interface .. 

12:35 <nginr> Can I use the load test tool for testing with http interface ? 

12:35 <roidrage> sure, i'm just saying, it's something to consider depending on use case 

12:36 <roidrage> nginr: yes, you can 

12:36 <nginr> Looks good. I I hope it doesnt take much time for me to start the test .. thanks