Hi there!
I try to reproduce the paper results for the freshness evaluation over PG 14 with streaming replication. As stated in the paper, PG-SR could have up to 5 seconds over SF10. However, I currently always got zero freshness. I tried to increase the ac and tc to 100 but had been in vain.
Did I do something wrong? I used the following benchmarking command with synchronous_commit="on":
./HATtrickBench -run -dsn hattrick1 -dsn2 hattrick2 -usr userName -pwd Password -wd 180 -td 360 -ac 100 -tc 100 -db postgres -t sp
Note that hattrick1 is the primary node and hattrick2 is the secondary node.
Any comments would be appreciated!
Thanks!
Hi there!
I try to reproduce the paper results for the freshness evaluation over PG 14 with streaming replication. As stated in the paper, PG-SR could have up to 5 seconds over SF10. However, I currently always got zero freshness. I tried to increase the ac and tc to 100 but had been in vain.
Did I do something wrong? I used the following benchmarking command with synchronous_commit="on":
./HATtrickBench -run -dsn hattrick1 -dsn2 hattrick2 -usr userName -pwd Password -wd 180 -td 360 -ac 100 -tc 100 -db postgres -t spNote that hattrick1 is the primary node and hattrick2 is the secondary node.
Any comments would be appreciated!
Thanks!