HBase for Real Time Analytics

[Ajay <aj...@gmail.com>]

Hi,

Can HBase be used or best fit for Real Time Analytics?. I went through
couple of benchmark between Cassandra Vs HBase (most of it done 3 years
ago). It mentioned that HBase is designed for intensive reads (but higher
latency for writes).

Secondly I want to understand how it is faster or support intensive reads
(or how it is better than Cassandra) though it is strongly consistent (CP).
In our case we may have high writes and reads (more reads though say 60%
reads). In that case is HBase a better fit?. We are  planning to use Spark
as the in memory computation engine.

Thanks
Ajay

Re: HBase for Real Time Analytics

[Ted Yu <yu...@gmail.com>]

Ajay:
See this thread: http://search-hadoop.com/m/DHED45Pavj1
where Andrew did some performance testing on 0.98.8 (the stable release)

The above would give you some idea of where HBase is performance-wise.

To my knowledge, there is no up-to-date comparison between recent releases
of Cassandra and HBase.

If read latency is of high priority to you, see HBASE-10070

Cheers

On Wed, Dec 17, 2014 at 8:41 PM, Ajay <aj...@gmail.com> wrote:
>
> Hi,
>
> Can HBase be used or best fit for Real Time Analytics?. I went through
> couple of benchmark between Cassandra Vs HBase (most of it done 3 years
> ago). It mentioned that HBase is designed for intensive reads (but higher
> latency for writes).
>
> Secondly I want to understand how it is faster or support intensive reads
> (or how it is better than Cassandra) though it is strongly consistent (CP).
> In our case we may have high writes and reads (more reads though say 60%
> reads). In that case is HBase a better fit?. We are  planning to use Spark
> as the in memory computation engine.
>
> Thanks
> Ajay
>