raid 0+1

[???_185]

is it same raid configuration betweeen 0+1 and 1+0?

[Srinivasa_6]

Yep - they both mean the same (Striped Mirroring) and they might be used interchangeably.

[Uwe Zessin]

I don't recall if there is any standard that describes what is what (striped mirrors vs. mirrored stripes). I never use 0+1 or 1+0, but always talk about SM or MS to avoid confusion.

[Srinivasa_6]

I stand corrected. In fact, there is a slight difference.

http://www.ofb.net/~jheiss/raid10/

[DragomirAtea]

Hi folks!

In my humble opinion there is a huge difference...

Mirrored stripes is dangerous to use!
They are infact two stripes that you mirror.

If you loose 1 diskdrive, the entire stripe will be lost, and youre running on a RAID 0. That means that you can only loose one disk in that configuration.

Stiped mirrors on the other hand, is multiple mirrors that you stripe. That means that you can loose 50% of your drives and still be up and running.

I would never set up a mirrored stripe in a working enviroment.

But..
Who says that I'm allways right, actually I was wrong once in 1976. (-;

Regards
Dragomir Zekic

[John Kufrovich]

Dragomir, your explanation is almost correct.

If you loose a drive from each set you loose access to your LUN.
With 0+1,
Drives 1 2 3 4
Set 1 = 0 0 0 0
Set 2 = 0 0 0 0

Set 1 is striped and Mirrored with Set 2. If you loose a drive in Set 1, say dr2. No problem you have set 2 for backup. BUT if you loose ANY drive in the Set 2. Then you have lost all access.

With 1+0
Drives 1 2 3 4
Set 1 = 0 0 0 0
Set 2 = 0 0 0 0

Each drive from each set is Mirrored with each other, then striped. SET 1 DR1 is Mirrored with SET 2 DR1. Then continue down the line. If we loose DR2 of set 1, no problem, we have the other drive2 as a mirror. The only time you could loose access if you lost drive2 of the Set 2. As Dragomir stated, you can loose 50% of your drives as long as the drive isn't the mirrored pair.

It is a matter of odds.
BTW, MSA and Smart Array Controllers do Raid 1+0.

[Basil Vizgin]

John,
how can I find what drives are in pair in raid 1 part of RAID1+0 array on SA (MSA) controllers?
If I split 1+0 array between 2 channels, are the members of pair on different channels?

This is important to understand how can array in MSA survive with enclosure failure.

[Shameer.V.A]

Hi zungwon ,
Please find the attached document. Hope this will help you to clarify..

Regards,

Shameer

[John Kufrovich]

Basil,

Take 6 drives, 3 from shelf A and 3 from shelf B. Disk A1 and B1 will be mirrored. A2 and B2, ...

[Basil Vizgin]

Thank you, John!

[dipesh_2]

Hi

There is a slight difference.

RAID 0+1 configuration where multiple disks are striped together into sets (sets A & B in the diagram, each set being as large as the resulting final volume), and then two or more sets are mirrored together.

RAID 1+0 configuration where two or more drives are mirrored together (mirrors 1-4 in the diagram), and then the mirrors (as many as are needed to result in the desired amount of space) are striped together.

[Sivakumar TS]

Hi Zung,

DEFENETLY IT IS NOT SAME

Simply

raid 0+1 ==> striping the mirrored volume

raid 1+0 ==> mirroring the striped volume

the redundancy differs in both config.
the mirror can survive a single disk failure and stripe cant.


With Regards,

Siva

[Uwe Zessin]

dipesh and Siva have exactly opposite definitions.

That's why I said I always talk about striped/mirrors or mirrored/stripes ;-)

[Basil Vizgin]

From MSA CLI Guide:

Note: If more than one pair of drives are included in a RAID 1 array, the data is
striped across the first half of the drives in the array and then each drive is mirrored to a
drive in the remaining half of the drives for fault tolerance. This method is referred to as
RAID 1+0.

[Adrian Parker]

Hi,

Using the CLI to create a 1+0 lun, will the MSA 1000 automatically mirror across enclosures if it can, or is the lun set-up based on the order of disks on the command line ?

Thanks,
Adrian

[Uwe Zessin]

It's my understanding that the MSA controller firmware is "smart" enough to use different enclosures on its own. I believe somebody from the MSA team said so some time ago here on ITRC, but I am not sure I could find the thread easily.

[Alzhy]

ZungWon,

Peace!

1. 0+1 and 1+0 both gives you the same capacity

2. AS far as redundancy - 0+1 (stripe and mirror) will be able to handle more disk failures.

3. Performance - it depends on where the RAIDing is done. If RAIDing is done on the host level say - using VxVM, and you are dealing with JBOD enclosures - one RAID scheme may be better than the other performance wise.

[John Kufrovich]

Nelson,

Please recheck your second statement.

RAID 1+0 offers more redundancy than 0+1.

RAID 0+1 will only handle a single disk failure in each mirrored set.

[Alzhy]

John you are absolutely right. RAID10 indeed is more redundant.

Take 8 disks, carve 4 mirror sets (RAID1), stripe(RAID0) accross this 4 mirror sets and you've RAID10. You can loose 1 disk from each mirror set and your stripe stays intact.

In VxVM, you have what's called layered Volumes. You can have a stripe of stripes, s stripe or mirrors, a stripe of RAID5's, etc.. further incresing the reliability and scalability of storage.

[Basil Vizgin]

Both 1+0 and 0+1 can handle same quantity of disk failures (for instance, second disk can fail in already failed half of 0+1 mirrorset). But probability of array failure with subsequentd HDD errors is much different (1/2 in RAID0+1 vs. 1/number of mirrors in RAID1+0).

[John Kufrovich]

Basil,
Your first sentence isn't correct.

RAID 01 can only handle two disk failures. One in each mirrored set.

RAID 10, can handle multiple disk failures as long as a failure isn't associated with an already failed mirrored member.

[Basil Vizgin]

John,
RAID0+1 of 10 disks (D):
---
DDDDD mirrored to DDDDD

STRIPE mirrored to STRIPE
---
Array lost one disk (X-failed drive), so secong set failed:
---
DDDDD mirrored to DDDXD
so
STRIPE mirrored to FAILED_STRIPE
---
But, as I understand, disks in failed stripe is still rotating and also can fail, for example:
---
DDDDD mirrored to DXDXD
so we have again
STRIPE mirrored to FAILED_STRIPE
---

Therefore in theory we can loose up to all disks of FAILED_SET.
Of course, workload in survived set is greater, so probability that next failed drive goes to working set is very hight.

[John Kufrovich]

Basil,
Your explanation is closer to a RAID 1+0 than 0+1.

Let's look at it in a different way.
With RAID 0+1, you have a
RAID 0 mirrored to a RAID 0.
What happens to a RAID 0 LUN when a drive is faulty? You loose that LUN. Same thing here, you loose that mirrored set.

With RAID 1+0, using 8 disk.
Drives 1 and 5 are mirrored (RAID1)
Drives 2 and 6 are mirrored (RAID1)
Drives 3 and 7 are mirrored (RAID1)
Drives 4 and 8 are mirrored (RAID1)
Then your data is stripped.
With this scenario, we could loose a single drive from each mirrored (RAID1)set and still function. In this case we can loose a max of 4.

[Basil Vizgin]

John,
In you example you assuming that next disk can fail only in active (survived) RAID0 set of RAID0+1, but by theory of probability it can fail in already failed RAID0.

[John Kufrovich]

You are correct, you could have a second failure in the failed RAID0. BUT, a failure is more likely to occur during I/O operations. The failed RAID0 will or should be marked BAD and no further I/O will occur.