Hi,
Before I start with my question, following are some references -
1) I want to keep track of creating and closing ceph connections properly so I am using singleton pattern for getting an instance of ceph.
2) following is kind of the steps-
a) creating a rados cluster
b) connecting to cluster
c) setting the IO context (pool)
and then using this object for all sorts of reads, write and updates.
Things have been fine for me as long as I was writing to one pool.
But due to design change I have to write to 2 separate pools. They both uses the same key, so basically to access the pool all I need to change is the io context.
Question - I decided to create the connection object and not assigning it to pool right away, rather deciding the pool name each time you want to read or write.
Will it add overhead on the same connection to switch between the pools ?
Or I should consider changing the design and creating 2 connection objects for 2 pools ?
Thanks a lot for your reply in advance,
Regards
Sonal Dubey
_______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
