Send Bristol mailing list submissions to
bristol@mailman.lug.org.uk
To subscribe or unsubscribe via the World Wide Web, visit
https://mailman.lug.org.uk/mailman/listinfo/bristol
or, via email, send a message with subject or body 'help' to
bristol-request@mailman.lug.org.uk
You can reach the person managing the list at
bristol-owner@mailman.lug.org.uk
When replying, please edit your Subject line so it is more specific
than "Re: Contents of Bristol digest..."
Today's Topics:
1. Re: SSDs (was Bristol Digest, Vol 588, Issue 2)
(Colin M. Strickland)
2. Re: SSDs (was Bristol Digest, Vol 588, Issue 2) (Alex Butcher)
----------------------------------------------------------------------
Message: 1
Date: Sun, 15 Mar 2015 09:57:32 +0000
From: "Colin M. Strickland" <cms@beatworm.co.uk>
To: "Bristol and Bath Linux User Group" <bristol@mailman.lug.org.uk>
Subject: Re: [bristol] SSDs (was Bristol Digest, Vol 588, Issue 2)
Message-ID: <7DB89AA3-748D-495E-9B1B-43544FDEFADE@beatworm.co.uk>
Content-Type: text/plain; format=flowed
On 3 Feb 2015, at 14:56, Alex Butcher wrote:
> On Tue, 3 Feb 2015, Fergus Allan wrote:
>
>> SSD's have been incredibly unreliable. Either dirty pages returned
>> or
>> complete failure.
>>
>> I've had 2 crucial SSD's, both failed totally at 3 months old; little
>> red
>> light comes on and no access to any part of the disk. The first also
>> returned dirty pages pretty much from day 1. I've haven't fixed all
>> my
>> corrupted data from that issue. That was a few years ago. We were
>> discussing ordering some SSDs at work and I suggested Crucial even
>> though I
>> don't trust their devices, expecting a positive reaction, but they
>> shot
>> that down as too unreliable.
>
> But of course you've got backups you can restore from, right?!?
>
> We make extensive use of SSDs in our centralised storage at work, with
> no
> particular problems. But, of course, we'll almost certainly be using
> enterprise kit rather than consumer stuff.
>
> When choosing an SSD for a new desktop, amongst the consumer stuff,
> the
> Samsung 850 PRO SSDs struck me as having appreciably better longevity
> ratings (150TBW for the 128GB and 256GB models, 300TBW for the 512GB
> and
> 1TB) than most others and with a warranty (10 years) to match.
>
> <http://www.storagereview.com/demystifying_ssd_endurance>
>
This popped into my INBOX today and seemed relevant to this thread
http://techreport.com/review/27909/the-ssd-endurance-experiment-theyre-all-dead
FWIW I've worked extensively with SSD storage alongside spinning disks
at moderate scale on and off for the last few years, as well as moving
to it exclusively on the desktop for the same period.
For my own "anecdata", I would say they're far less prone to sudden
catastrophic failure than traditional HDD, across my sample set of a
couple of hundred devices. For both classes of device, I would posit you
pretty much get what you pay for. They work *differently* from disks,
I suggest you read the numbers carefully, and analyse your own typical
I/O patterns if you actually are looking to improve performance.
Heuristics like 'they're faster but not as reliable', 'swap isn't
useful' etc. will not get you very far. Optimising things you haven't
even measured is a bit of a pantomime.
There's an awful lot of knobs you can twiddle to tune wrt disk and
filesystem performance, if you do care - OTOH glom everything together
into one LVM RAID volume built out of drives with the best reliability
guarantees is far more convenient.
--
Regards,
Colin M. Strickland, cms, 'that guy'.
------------------------------
Message: 2
Date: Sun, 15 Mar 2015 11:23:38 +0000 (GMT)
From: Alex Butcher <lug@assursys.co.uk>
To: Bristol and Bath Linux User Group <bristol@mailman.lug.org.uk>
Subject: Re: [bristol] SSDs (was Bristol Digest, Vol 588, Issue 2)
Message-ID:
<alpine.LRH.2.11.1503151114230.18925@zlgugi.of5.nffheflf.cev>
Content-Type: text/plain; charset="utf-8"; Format="flowed"
On Sun, 15 Mar 2015, Colin M. Strickland wrote:
> On 3 Feb 2015, at 14:56, Alex Butcher wrote:
>> When choosing an SSD for a new desktop, amongst the consumer stuff, the
>> Samsung 850 PRO SSDs struck me as having appreciably better longevity
>> ratings (150TBW for the 128GB and 256GB models, 300TBW for the 512GB and
>> 1TB) than most others and with a warranty (10 years) to match.
>>
>> <http://www.storagereview.com/demystifying_ssd_endurance>
>>
>
> This popped into my INBOX today and seemed relevant to this thread
>
> http://techreport.com/review/27909/the-ssd-endurance-experiment-theyre-all-dead
>
> FWIW I've worked extensively with SSD storage alongside spinning disks at
> moderate scale on and off for the last few years, as well as moving to it
> exclusively on the desktop for the same period.
>
> For my own "anecdata", I would say they're far less prone to sudden
> catastrophic failure than traditional HDD, across my sample set of a couple
> of hundred devices.
I'll give credit to your large-ish sample set (larger than my own, at least
- a mere single SSD, so far).
However:
"The Corsair, Intel, and Kingston SSDs all issued SMART warnings before
their deaths, giving users plenty of time to preserve their data. The
HyperX's warnings ended up being particularly premature, but that's better
than no warning at all. Samsung's own software pronounced the 840 Series and
840 Pro to be in good health before their respective deaths. Worryingly, the
840 Series' uncorrectable errors didn't change that cheery assessment.
If you write a lot of data, keep an eye out for warning messages, because
SSDs don't always fail gracefully. Among the ones we tested, only the Intel
335 Series and first HyperX remained accessible at the end. Even those
bricked themselves after a reboot. The others were immediately unresponsive,
possibly because they were overwhelmed by incoming writes before attempted
resuscitation."
<http://techreport.com/review/27909/the-ssd-endurance-experiment-theyre-all-dead/4>
Hence my warning to back up SSDs more dilligently than (many of us) do our
spinning rust HDDs. Of course, if you already dilligently back up your
HDDs, then SSDs indeed don't need to be treated any differently.
I'm beginning to regret my decision to consider Samsung again after previous
bad experiences with their (consumer) products put them on my "never again"
list. Not to mention:
"In October, Samsung released a tool to address a slowdown in 840 EVO
Sequential Read speeds reported by a small number of users after not using
their drive for an extended period of time. This tool effectively and
immediately returned the drive?s performance to normal levels,? the company
told AnandTech in an email. ?We understand that some users are experiencing
the slowdown again. While we continue to look into the issue, Samsung will
release an updated version of the Samsung SSD Magician software in March
that will include a performance restoration tool."
<http://www.maximumpc.com/samsung_promises_another_fix_840_evo_ssd_performance_issues900>
(which I'm unaffected by, but it's indicative of poor testing by Samsung).
Best Regards,
Alex
------------------------------
_______________________________________________
Bristol mailing list
Bristol@mailman.lug.org.uk
https://mailman.lug.org.uk/mailman/listinfo/bristol
End of Bristol Digest, Vol 593, Issue 4
***************************************
Tidak ada komentar:
Posting Komentar