I was curious to see how postgres would perform with wal on a tmpfs vs disk here are some numbers I got from pgbench.
Letme know if I did something stupid, this is the first time I've used pgbench. The wal on tmpfs method is not
significantlyfaster. <br /><br />[[ WAL ON TMPFS ]]<br />pgbench -i -s 10 -U postgres -d benchmark<br />...<br
/>pgbench-Upostgres -s 10 -c 10 -t 10000 benchmark<br />starting vacuum...end.<br />transaction type: TPC-B (sort
of)<br/>scaling factor: 10<br />number of clients: 10 <br />number of transactions per client: 10000<br />number of
transactionsactually processed: 100000/100000<br />tps = 5817.693724 (including connections establishing)<br />tps =
5825.646441(excluding connections establishing)<br /><br />[[ WAL ON EXT2 14 U320 DRIVE RAID10 WITH BBU (same as data)
]]<br/>pgbench -Upostgres -s 10 -c 10 -t 10000 benchmark<br />starting vacuum...end.<br />transaction type: TPC-B (sort
of)<br/>scaling factor: 10<br />number of clients: 10 <br />number of transactions per client: 10000<br />number of
transactionsactually processed: 100000/100000<br />tps = 5653.187997 (including connections establishing)<br />tps =
5660.554438(excluding connections establishing)<br /><br />pgbench -Upostgres -s 100 -c 10 -t 10000 benchmark<br
/>startingvacuum...end.<br />transaction type: TPC-B (sort of)<br />scaling factor: 100<br />number of clients: 10<br
/>numberof transactions per client: 10000<br />number of transactions actually processed: 100000/100000 <br />tps =
5536.019864(including connections establishing)<br />tps = 5543.834350 (excluding connections establishing)<br /><br />
