Win10 Disk Managment software or Gigabyte Motherboard BIOS to implement RAID 1?

[Imacflier]

As a result of my system giving itself an internal shower (enema?) as a result of a leaking hose in my now discarded AIO cooler, I am rebuilding my system.

I have two 500GB Samsung EVO 850 SSD's and a 480GB Crucial SSD. I plan on setting up RAID 1 using the two Samsung SSD's and using the Crucial SSD as an external USB3 backup run through Acronis 2018 automatically one a week....or when software is added or updated.

I solicit comments and suggestions concerning this plan.

Also, should I use the Win10 Disk Manager or use the Gigabyte BIOS firmware to implement RAID 1?

Similar questions on the forum have resulted in confusing me and I hope a lively discussion will take place so that I learn enough to avoid a disaster!

Thanks In Advance,
Larry

 

[Snipergod87]

If you want your RAID 1 to be bootable it needs to be through the BIOS. I personally have found RAID 1 on SSD's to be unnecessary as their reliability is much higher (though I do still use RAID 1 for mission critical servers)

 

[Imacflier]

Snipergod87,

I thought their reliability to be sufficient, too. When wetted whilst operating, however, the seem to have the same reliability as dropping your cellphone into your drink!

Thank you about the need to implement via the BIOS....I had no idea.

Larry

 

[USAFRet]

There are much easier, better, ways to safeguard your data vs a full RAID 1.

 

[Snipergod87]

You will need to change your SATA mode from AHCI to IDE, and then during boot press Control+I to open the Intel Matrix Raid Utility, when when you reinstall Windows you will need to point it to the RAID drivers if it can't see the disks (they can be downloaded from Gigabyte or Intels website, they are sometimes called F6Floppy.zip)

 

[androbourne]

I'd say for sure to avoid Microsoft Windows software Raid as much as possible. Raid is lost if OS is corrupted and unrecoverable.

The next best option is, as others stated with internal BIOS raid

But if you are doing this for performance reasons. Then I would recommend a true Raid card over the motherboard BIOS raid.

And like it was mentioned. Doing it for only 2 SSDs isn't really worth it compared to the hassle of Raid management vs benefits. If you are going to do a Raid for performance. I'd suggest doing it right and adding more then just 2 drives to the Raid along with a true Raid card.

 

[USAFRet]

Snipergod87,

I thought their reliability to be sufficient, too. When wetted whilst operating, however, the seem to have the same reliability as dropping your cellphone into your drink!

Thank you about the need to implement via the BIOS....I had no idea.

Larry

You seem to be doing this to ward off a dead drive, because the last one took a bath and drowned.

All that requires is a good backup situation, rather than a RAID 1.
If you had this RAID 1 in place when the cooler peed all over...both drives would have taken a bath, killing both of them.

A scheduled image, off to a whole other drive (external?) would not have been affected in this bath issue.
Simply slot in a new drive, and recover the image from last night that was automatically done.
Easy, and it 'just works'.

 

[androbourne]

^ Exactly.

I would personally recommend an external drive and Veeam Agent for Windows. It's by far the best free program I've used and I even implemente the business edition in enterprise environments almost on a daily bases.

 

[Imacflier]

Thanks all. The point of using Raid 1 PLUS an external backup, is avoiding losing my between backups work product when (not if) I lose one of the SSD's.

Since have been doing an every other day manual clone from SSD1 to SSD2 prior to the water damage, it seemed to me that using RAID 1 would accomplish the same thing only easier and better.

For some reason this whole subject seems to get folks awfully 'het up'. So tell me, given that I have three five hundred class SSD's and 15TB of rotating drive space (contents of the 15 TB are replaceable media files, btw). and given that I am concerned about losing work product BETWEEN external backups...What would you all recommend as an alternative to what I have proposed??

Larry

 

[USAFRet]

A good imaging set up also wards off the other, far more common, forms of data loss.
Malware, accidental deletion, ransomware, etc, etc.

A RAID 1 simply means it happens on two physical drives at the same moment.

 

[androbourne]

Thanks all. The point of using Raid 1 PLUS an external backup, is avoiding losing my between backups work product when (not if) I lose one of the SSD's.

Since have been doing an every other day manual clone from SSD1 to SSD2 prior to the water damage, it seemed to me that using RAID 1 would accomplish the same thing only easier and better.

For some reason this whole subject seems to get folks awfully 'het up'. So tell me, given that I have three five hundred class SSD's and 15TB of rotating drive space (contents of the 15 TB are replaceable media files, btw). and given that I am concerned about losing work product BETWEEN external backups...What would you all recommend as an alternative to what I have proposed??

Larry

Well real question, what is the amount of time back you are willing to lose data at any given time?

If one day is too long. (aka waiting for backups to run, once daily). Then you may need a more active solution. Such as a Cloud Storage backup that runs continuous as files are changed. Programs like Carbonite or Code42, do just that.

You could also adjust backup times to your external to hourly or something similar, but that may bog down your system. Of course that depends on your specs, and how much data has change in said hour. It should work just fine for your needs. Give it a try. I personally have not had issues with Veeam on an hourly schedule. But I have a pretty beefy system.

 

[USAFRet]

I have my main system scheduled to write an image (or incremental depending on the day), off to my large NAS every night at 1AM.

5 x SSD, each with different purposes. ~2TB total.
Writing to a 10.6TB spinning drive NAS. That NAS holds other stuff, but not relevant here.
Each individual drive gets its own Image, in its own folder.

In case of a singular dead drive, simply recover from last nights backup.
If ALL of them die at the same time, write each to individual partitions on a large spinning drive for temporary ops.

The backups are never more than 24 hours out of date.
And, each individual drive, or the whole system, can be reconstituted to any state in the last 14 days.
Something bad happened last Tuesday, and I did not notice until Friday? No prob. Recover from Monday nights backup.

Also, I could open up any individual Full or Incremental image, and recover a single file.
I need the copy of my resume as it was on Wednesday? No prob.

A full clone, or a RAID 1, gives up a lot of possible functionality.

I use Macrium Reflect. Paid version on my main system, free version on all the others in the house.
I've heard good things about the above mentioned Veeam, but I've never used it.

 

[androbourne]

Yeah I've worked with Macrium Reflect before as well. It's pretty comparable to Carbonite.

I run Veeam backups hourly on my main system and daily on the house items to a Netgear 6TB NAS. I then use built in NAS software to backup from the NAS to external drives. I have two large externals that I rotate on a weekly bases.

But for some that is pretty extreme. To be fair I also host websites and gaming servers so on point backups are pretty critical.