kovacik

3 ms·

1. 1.▲
   by kovacik 9y ago

   I actually did try this and the results were worse. I did not investigate it further and went with primarycache=all
2. 2.▲
   by kovacik 9y ago

   I did set the recommended 8k block size for PostgreSQL dataset as described in the blogpost. Also note the logbias=throughput option.
3. 3.▲
   by kovacik 9y ago

   The size of postgresql database was ~74GB for every test combination. The bare metal server had 256GB RAM for every test combination. The only difference was the amount of RAM PostgreSQL could use - as specified in postgresql.conf. So, 74GB
4. 4.▲
   by kovacik 9y ago

   I think it's pretty fair to test each OS with its filesystem of choice. I'm aware that you can use ZFS on Linux, but I'm not (yet) brave enough to recommend ZFS+Linux. And yes there's btrfs but would you trust it with yo
5. 5.▲
   by kovacik 9y ago

   Before the benchmark I did some test with FreeBSD+ZFS with LZ4 on/off. The LZ4 CPU overhead was pretty negligible and performance was slightly better than without compression, due to lower IO usage. That's the reason I've cho
6. 6.▲
   by kovacik 9y ago

   There is a 30 min warmup period before the actual benchmark. It is shown in the benchmarking script.
7. 7.▲
   by kovacik 9y ago

   Hello, OP here. I'm certain that you can fine tune every OS for specific use case. I may indeed do that in a future blogpost. The question is what to compare ? Should I compare Linux kernel versions, PostgreSQL versions, filesystems (a