Blake,
The two instances that are "optimized" are currently setup like the below:
default_statistics_target = 100
maintenance_work_mem = 512MB
constraint_exclusion = on
checkpoint_completion_target = 0.9
effective_cache_size = 40GB
work_mem = 64MB
wal_buffers = 8MB
max_connections = 100
shared_buffers = 16GB
join_collapse_limit = 9
The server itself has the following that are relevant to Pg performance:
vm.overcommit_memory = 2
vm.nr_hugepages = 16893
As I said, I'm considering changing the configuration so that the
database instance being tested gets all of the resources, and I may shut
the others down individually.
As for Pg10, we're using it on our training server and on this test
server. It's one of the two currently "optimized." I have not
encountered any issues with Evergreen and Pg 10. The performance is
comparable to Pg 9.6 so far.
Cheers,
Jason
_______________________________________________
Evergreen-dev mailing list
Evergreen-dev@list.evergreen-ils.org
http://list.evergreen-ils.org/cgi-bin/mailman/listinfo/evergreen-dev
