Received: from malur.postgresql.org ([217.196.149.56]) by arkaria.postgresql.org with esmtp (Exim 4.80) (envelope-from ) id 1ZC9jK-00067x-Jn for pgsql-performance@arkaria.postgresql.org; Mon, 06 Jul 2015 16:58:54 +0000 Received: from localhost ([127.0.0.1] helo=postgresql.org) by malur.postgresql.org with smtp (Exim 4.84) (envelope-from ) id 1ZC9jK-0005wX-36 for pgsql-performance@arkaria.postgresql.org; Mon, 06 Jul 2015 16:58:54 +0000 Received: from magus.postgresql.org ([2a02:c0:301:0:ffff::29]) by malur.postgresql.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_CBC_SHA384:256) (Exim 4.84) (envelope-from ) id 1ZC9gv-0003rM-FM for pgsql-performance@postgresql.org; Mon, 06 Jul 2015 16:56:25 +0000 Received: from cerberus.pinpointresearch.com ([66.7.238.130] helo=polaris.pinpointresearch.com) by magus.postgresql.org with esmtp (Exim 4.84) (envelope-from ) id 1ZC9gr-0003fk-0B for pgsql-performance@postgresql.org; Mon, 06 Jul 2015 16:56:24 +0000 Received: from [192.168.1.179] (betelgeuse.pinpointresearch.com [192.168.1.179]) by polaris.pinpointresearch.com (Postfix) with ESMTP id 514FEE00EC83; Mon, 6 Jul 2015 09:56:18 -0700 (PDT) Message-ID: <559AB332.50304@pinpointresearch.com> Date: Mon, 06 Jul 2015 09:56:18 -0700 From: Steve Crawford User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Thunderbird/31.7.0 MIME-Version: 1.0 To: "Wes Vaske (wvaske)" , "pgsql-performance@postgresql.org" Subject: Re: New server: SSD/RAID recommendations? References: <31701C89A44B714FB8D242F806E7286C5AED32FA@NTXBOIMBX01.micron.com> In-Reply-To: <31701C89A44B714FB8D242F806E7286C5AED32FA@NTXBOIMBX01.micron.com> Content-Type: multipart/alternative; boundary="------------000700090509030407060904" X-Pg-Spam-Score: -2.6 (--) List-Archive: List-Help: List-ID: List-Owner: List-Post: List-Subscribe: List-Unsubscribe: X-Mailing-List: pgsql-performance Precedence: bulk Sender: pgsql-performance-owner@postgresql.org This is a multi-part message in MIME format. --------------000700090509030407060904 Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: quoted-printable On 07/02/2015 07:01 AM, Wes Vaske (wvaske) wrote: > > What about a RAID controller? Are RAID controllers even available for=20 > PCI-Express SSD drives, or do we have to stick with SATA if we need a=20 > battery-backed RAID controller? Or is software RAID sufficient for SSD=20 > drives? > > Quite a few of the benefits of using a hardware RAID controller are=20 > irrelevant when using modern SSDs. The great random write performance=20 > of the drives means the cache on the controller is less useful and the=20 > drives you=E2=80=99re considering (Intel=E2=80=99s enterprise grade) wi= ll have full=20 > power protection for inflight data. > For what it's worth, in my most recent iteration I decided to go with=20 the Intel Enterprise NVMe drives and no RAID. My reasoning was thus: 1. Modern SSDs are so fast that even if you had an infinitely fast RAID=20 card you would still be severely constrained by the limits of SAS/SATA.=20 To get the full speed advantages you have to connect directly into the bu= s. 2. We don't typically have redundant electronic components in our=20 servers. Sure, we have dual power supplies and dual NICs (though=20 generally to handle external failures) and ECC-RAM but no hot-backup CPU=20 or redundant RAM banks and...no backup RAID card. Intel Enterprise SSD=20 already have power-fail protection so I don't need a RAID card to give=20 me BBU. Given the MTBF of good enterprise SSD I'm left to wonder if=20 placing a RAID card in front merely adds a new point of failure and=20 scheduled-downtime-inducing hands-on maintenance (I'm looking at you,=20 RAID backup battery). 3. I'm streaming to an entire redundant server and doing regular backups=20 anyway so I'm covered for availability and recovery should the SSD (or=20 anything else in the server) fail. BTW, here's an article worth reading:=20 https://blog.algolia.com/when-solid-state-drives-are-not-that-solid/ Cheers, Steve --------------000700090509030407060904 Content-Type: text/html; charset=utf-8 Content-Transfer-Encoding: quoted-printable
On 07/02/2015 07:01 AM, Wes Vaske (wvaske) wrote:

What about a RA= ID controller? Are RAID controllers even available for PCI-Express SSD drives, or do we have to stick with SATA if we need a battery-backed RAID controller? Or is software RAID sufficient for SSD drives?

=C2=A0

Quite a few of the benefits of using a hardware RAID controller are irrelevant when using modern SSDs. The great random write performance of the drives means the cache on the controller is less useful and the drives you=E2=80=99re consi= dering (Intel=E2=80=99s enterprise grade) will have full power prote= ction for inflight data.


For what it's worth, in my most recent iteration I decided to go with the Intel Enterprise NVMe drives and no RAID. My reasoning was thus:

1. Modern SSDs are so fast that even if you had an infinitely fast RAID card you would still be severely constrained by the limits of SAS/SATA. To get the full speed advantages you have to connect directly into the bus.

2. We don't typically have redundant electronic components in our servers. Sure, we have dual power supplies and dual NICs (though generally to handle external failures) and ECC-RAM but no hot-backup CPU or redundant RAM banks and...no backup RAID card. Intel Enterprise SSD already have power-fail protection so I don't need a RAID card to give me BBU. Given the MTBF of good enterprise SSD I'm left to wonder if placing a RAID card in front merely adds a new point of failure and scheduled-downtime-inducing hands-on maintenance (I'm looking at you, RAID backup battery).

3. I'm streaming to an entire redundant server and doing regular backups anyway so I'm covered for availability and recovery should the SSD (or anything else in the server) fail.

BTW, here's an article worth reading: https://blog.algolia.com/when= -solid-state-drives-are-not-that-solid/

Cheers,
Steve

--------------000700090509030407060904--