Interesting topic and we have a JIRA already raised for such a feature.
But still the work is in progress
On Tue, Apr 9, 2013 at 10:07 PM, Stack <[EMAIL PROTECTED]> wrote:
> On Tue, Apr 9, 2013 at 4:41 AM, Bing Jiang <[EMAIL PROTECTED]>
> > hi,
> > There are some physical machines which each one contains a large ssd(2T)
> > and general disk(4T),
> > and we want to build our hdfs and hbase environment.
> What kind of workload do you intend to run on these machines? Do you have
> enough space running all of your work load on SSD? At an extreme, you
> could have two clusters -- one running on SSDs for low latency workloads
> and the other on spinning disk -- and perhaps your segregation is such that
> having to copy between the two systems is rare, etc., etc.