[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: PostgreSQL+ZFS

Atif CEYLAN put forth on 12/31/2010 4:58 PM:
> Hi all,
> I have a large postgresql database system and I want to migrate to a new
> and fast storage system (10 Gbp/s FC network). But 150x3 ssd disk on my
> db server and I want to use ext4 file system (raid5) at the ssd disks as
> xlog storage or use zfs (raidz) as disk buffer cache.
> What is your idea?

In as few words as possible?  You're a nut job.  The mere mention of
using FUSE ZFS in any production context on Linux proves it.  As does
mentioning running RAID 5 on 3 SSDs, or any SSDs for that matter, or
running RAID 5 in a db context, for that matter.

What exactly are you really asking for advice on?  You've left out the
most important detail:  what your database application is, and what you
actually do with it.

What type of data does your db house?
How much data?  Total GB?
Are you currently short of space?
Are you currently short of IOPS capacity?
How many concurrent transactions?
What types of transactions?
Is it read only/heavy or transactional like point of sale?

What is a "large postgresql database system"?  What exactly do you mean
by this?  Does large mean heavy transaction load?   Or does it simply
mean lots of data housed?  Or is it simply BS?

Giving any relevant recommendations WRT SAN, SSD, or filesystem
performance is meaningless at this point when you've given exactly zero
details about your database workload.  The workload drives all other
aspects of the system design.  For all we know your db could simply
contain your pirated music collection and would be completely
comfortable on a single 1TB SATA disk.

Give us the pertinent details and we can probably give you some decent
"ideas", as you requested.


Reply to: