| From: | Claudio Freire <klaussfreire(at)gmail(dot)com> |
|---|---|
| To: | Igor Chudov <ichudov(at)gmail(dot)com> |
| Cc: | pgsql-performance(at)postgresql(dot)org |
| Subject: | Re: Postgres for a "data warehouse", 5-10 TB |
| Date: | 2011-09-11 14:44:18 |
| Message-ID: | CAGTBQpYyzdGKwQ12vYjv-neFmYfw1S899+KPAvSwkhoFdz-3JA@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-performance |
On Sun, Sep 11, 2011 at 4:21 PM, Igor Chudov <ichudov(at)gmail(dot)com> wrote:
> Quantitatively, what would you say is the write speed difference between
> RAID 10 and RAID 6?
https://support.nstein.com/blog/archives/73
There you can see a comparison with 4 drives, and raid 10 is twice as fast.
Since raid 5/6 doesn't scale write performance at all (it performs as
a single drive), it's quite expected. 12 drives would probably be
around 6 times as fast as raid 6.
You definitely should do some benchmarks to confirm, though.
And Andy is right, you'll have a lot less space. If raid 10 doesn't
give you enough room, just leave two spare drives for a raid 0
temporary partition. That will be at least twice as fast as doing
temporary tables on the raid 6.
You'll obviously have to get creative, tons of options.
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Marti Raudsepp | 2011-09-11 17:02:06 | Re: Postgres for a "data warehouse", 5-10 TB |
| Previous Message | Andy Colson | 2011-09-11 14:36:25 | Re: Postgres for a "data warehouse", 5-10 TB |