[Bigdata-developers] Namespace management

[Stas M. <sma...@wi...>]

Hi!

I am looking at the following scenario: I have a large (~70G now)
database, in which most of the data is stored in specific namespace. I
want to reload the database with fresh data, but it takes a lot of time
(several days) and I don't want to take the database down for it. So I
was thinking, maybe it is possible to load the data into the secondary
namespace, and then shut down the database for a short time, remove the
old namespace and put the new one instead. So my questions are:

1. Is this a good idea in general or the same can be better achieved in
other ways?
2. Can namespaces be renamed/moved?
3. If I delete a huge namespace, would it have performance impact?
4. if we're talking about running it on the same machine, would it be
better to have one Blazegraph instance working on two namespaces in
parallel, or maybe two Blazegraph instances working on separate files
(then the question of namespace management becomes moot)?

Thanks,
-- 
Stas Malyshev
sma...@wi...