hbase-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Nicolas Spiegelberg <nspiegelb...@fb.com>
Subject Re: Interesting YCBS Benchmark
Date Fri, 11 Feb 2011 17:01:11 GMT
These slides were presented at FOSDEM, the conference where I just spoke about FB & HBase.
 I went to the talk.   

There were some glaringly obvious problems that came out during the Q&A.  Someone asked
how many regions he had. Didnt know. Are you network or IO or CPU bottlenecked? Didn't know.
 Too many of the results for HBase and other DBs, he just said "I'm not sure what's going
on here". 

The publisher was a grad student who seemed to have done this work for his masters.  I'm glad
we have ppl interested in analyzing HBase, but you need to understand that this is not a thorough
evaluation. The main thing for him was learning how all these DBs worked at a high level and
thinking of novel benchmark scenarios.  

Sent from my iPhone

On Feb 11, 2011, at 7:32 AM, "Stack" <stack@duboce.net> wrote:

> That paper looked like complete nonsense.  The slide on HBase has
> misunderstandings about how we work.   The data size is miniscule.
> We get slower when you add nodes (?).
> 
> St.Ack
> 
> On Fri, Feb 11, 2011 at 7:01 AM, David Engfer <david.engfer@gmail.com> wrote:
>> http://nosql.mypopescu.com/post/3163240962/ycbs-benchmark-results-for-cassandra-hbase-mongodb
>> 
>> Not sure what the issue was there with HBase, but would this be much
>> improved if they tried it again with 0.90?
>> 
>> Thanks,
>> _______________________________
>> David Engfer
>> [817.360.4923]
>> david.engfer@gmail
>> http://www.engfers.com
>> 

Mime
View raw message