Hi Patric,
I've spent some time looking for replication and load balancing
software for Postgresql. The available add ons or proxies don't
seem to be what I was looking for.
Well sequoia seems.
I read most things I could find, but there are still some open
questions?
Is it right that the latest sequoia can talk to postgresqlserver
without jdbc-postgres driver?
No. The client application can talk to Sequoia without jdbc-postgres
driver (using Myosotis) but Sequoia will still use jdbc-postgres driver
to connect to the Postgresql database.
Is it right that the latest sequoia supports ODBC without using carob?
Yes, if you use Myosotis, it will intercept the native Postgres protocol
and convert it into Sequoia protocol. So if you use the Postgres ODBC
Driver, it can be plugged directly into Myosotis.
Is it planned supporting PG directly like libmysequoia for Postgresql?
I don't know if Continuent is going to work on such development. What
API would you like so see supported?
So for me as php user, I've to use the odbc driver?
Unfortunately our framework makes use of the async query execution
feature offered by postgres, to massively execute queries concurrently
to fetch data faster and optimize build time, suppose there is no way
using this with sequoia?
I read somewhere that uni-cluster supports postgres natively?
Does this mean, on client site or DB side?
This means on the client side. This is basically Sequoia with Myosotis.
You can use any Postgres driver (ODBC or any other lib) and Myosotis
will intercept the Postgres protocol. You will not be able to have
transparent failover at the application level but all other features
should work without application changes.
Is there any proof of robustness of sequoia, e.g. well known
production web site with heavy load?
I think this is a recurrent question that no one wants to answer. I am
under NDA with all customers I have been working with. So I can't tell
but I hope some people will step up! You can see some references on
Continuent web site but very big sites (well known) are usually running
Oracle RAC and don't even bother looking at solutions like Sequoia.
Does sequoia support enhanced features like dblink?
e.b SELECT * from dblink()... or is this incompatible with sequoia
or unicluster?
There is no specific support for dblink in Sequoia, not sure about the
latest unicluster. I don't know this extension but it looks like this
is used for persistent connections. Sequoia supports persistent
connections (look at the Sequoia driver options).
What specific use are you making of dblink?
Thanks for your interest in Sequoia,
Emmanuel
--
Emmanuel Cecchet
FTO @ Frog Thinker
Open Source Development & Consulting
--
Web: http://www.frogthinker.org
email: [EMAIL PROTECTED]
Skype: emmanuel_cecchet
_______________________________________________
Sequoia mailing list
[email protected]
https://forge.continuent.org/mailman/listinfo/sequoia
