[pgpool-general: 1469] [Benchmarks] Replication mode VS Master/Slave mode

[Thomas Martin]

Hello everyone.

Does someone already did some benchmarks to compare Replication Mode
with Master/Slave_mode (using streaming replication) ?

I did some benchmarks on pgpool-II using pgbench-tools
(https://github.com/gregs1104/pgbench-tools) and I was really
surprised by results.
Indeed with my tests the Master/Slave mode was more or less twice
better than the replication mode (obviously using the same servers and
sames postgresql configurations during both tests).

I used pgbench-tools to test simple select or more complicated
transactions: insert, update, select and in both case  Master/Slave
was really better.

I can't figure out why there is such a huge difference.
I would love to have some feedbacks about that kind of experience.


Thanks.

Thomas