Re: Catching up Production from Warm Standby after maintenance - Please help - Mailing list pgsql-admin

From Kevin Grittner
Subject Re: Catching up Production from Warm Standby after maintenance - Please help
Date
Msg-id 4A547A770200002500028580@gw.wicourts.gov
Whole thread Raw
In response to Re: Catching up Production from Warm Standby after maintenance - Please help  (Scott Marlowe <scott.marlowe@gmail.com>)
List pgsql-admin
Scott Marlowe <scott.marlowe@gmail.com> wrote:
> Jennifer Spencer<jenniferm411@hotmail.com> wrote:

>> new ones coming have 5 TB each of RAID disks mirrored across to
>> another 5TB (20 5.4gb spinning disks).  Not sure the RAID level but
>> I think it's level six.  They each have 16 CPUs, and (I believe)
>> 128 Gb of RAM, connected to our NFS network on a 10g switch (at
>> least I think it's the 10 gigabit switch - it might be one
>> gigabit).

> get a battery backed cache if you can.

I'd say that with the rest of the hardware described, performance will
be very disappointing without a good RAID controller with battery
backed cache configured for write-back.  Also, file system (we use
XFS) and mount options (we use noatime and nobarrier), and elevator
(we use deadline) can make a big difference.

-Kevin

pgsql-admin by date:

Previous
From: Scott Marlowe
Date:
Subject: Re: Catching up Production from Warm Standby after maintenance - Please help
Next
From: "Kevin Grittner"
Date:
Subject: Re: Catching up Production from Warm Standby after maintenance - Please help