• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Reliability 2 HDs in RAID1 or 1 SSD?

K

klau1

Member
Between a RAID 1 setup of 2 HDs in RAID 1 or 1 Sandforce/Indilinx SSD, what is more reliable?

Application: boot drive for a Windows Server 2003 domain server @ home.
 
A RAID 1 array of spindle hard drives is pretty well established and is pretty good. I've never seen an entire RAID 1 array lost, but that doesn't mean it can't happen.

The reliability of SSDs is still a question mark in my mind. I've seen several folks on these Forums who've had their SSDs fail.

Of course, a RAID 1 array is also capable of scrambling data. I've had my Exchange Server's RAID 1 database damaged twice in six years by power outages (and, yeah, I had a UPS). Backups, not RAID, are what saved the day.
 
Last edited:
How did that happen? Did the UPS fail to shutdown the PC before it ran out of juice? Have you considered equipping your RAID card with an onboard battery?

Reportedly, from reviews, SSDs are likely to remain readable after failure.

Surely it depends on which component fails, the NANDs (which case, the existing data is readable/recoverable) or the controller itself (which case nothing is recoverable).

When SSDs fail, how likely will existing data remain readable?
 
klau1 said:
How did that happen? Did the UPS fail to shutdown the PC before it ran out of juice? Have you considered equipping your RAID card with an onboard battery?
Click to expand...
The first time, the UPS batteries were dying and the UPS didn't give a warning. The second that the power failed, the UPS cut off.

The second time, I was in the process of shutting down the server, but wanted to silence the alarm because somebody was trying to sleep. I was tired, and turned off the UPS power switch before the server had shut down, scrambling a couple of databases.

Stuff happens.

That was my personal SBS 2003 server, which used an onboard Dell CERC RAID. No BBU option.
 
Last edited:
Since an SSD is so small but performs so well for the OS, and an HDD is so cheap for a large capacity, I think the best and most reliable way would be to simply image the SSD to a HDD. This gives the speed of the SSD, and the redundancy of two systems for a reasonable cost.
 
be careful some ssd's enable write cache.

the SUA2200XL is a nice unit - it does a self test once a week by default to ensure you aren't going to lose power. if you can score two with Rendundant power supplies you can know when it's not cutting it - and even add a battery that extends 40% runtime from 40 to 400 minutes.

raid-1 should only be used with proper drives (RE3/RE4 WD for instance) otherwise you will probably meet no operating system found sooner than not.

imo on-board raid (intel matrix or nvidia) is best for raid-0 - which is not a bad idea - with proper backups. It's really not that hard to do like brushing your teeth and a habit that will pay back everywhere you go (preach it)
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top