Two-way Mirror and Three-Way Mirror

Question

Hi

Please I need help.

I have 2 questions:

first question:

When selecting a layout either (Simple, Mirror, or Parity), after it, it creates a virtual disk and volume. A column is created, but I don't know in which sense does it mean, where is it, and how it works in the background?

second question:

I run Windows server 2022 and I have done lab for Two-way mirror and Three-way mirror, in both cases I use 11 Disk with size 5GB.

in Two-way mirror: as I understand, 2 disks will keep running after a single disk failure, however, after increasing the disk to 6 the disk failure increased and become 2.

why did this happen and how does it work in the background

3 Disks, max fail 1

4 Disks, max fail 1

5 Disks, max fail 1

6 Disks, max fail 2

7 Disks, max fail 2

8 Disks, max fail 3

the same thing goes for a Three-way mirror: 3 disks will keep running after 2 disk failures, after I increased the disk to 10 the disk failure increased and become 3, also when I reached 11 disks the disk failure increased and become 4

5 Disks, max fail 2

6 Disks, max fail 2

7 Disks, max fail 2

8 Disks, max fail 2

9 Disks, max fail 2

10 Disks, max fail 3

11 Disks, max fail 4

Please in both questions if there are pictures that describe how it works then it would be really helpful.

kind regards,

Answer 1

In Storage Spaces, a column is effectively a stripe of data. A virtual disk created with n columns will write data across n physical disks simultaneously.

Increasing the number of columns improves read and wtite performance, at the expense reducing flexibility when expanding the pool. For example, if a particular virtual disk (having n columns) runs out of storage, you'll need to expand the pool by a minimum of n physical disks to accommodate the virtual disk's layout.

As for understanding 2-way and 3-way fault tolerance, perhaps this article will help. The discussion refers to Azure Stack HCI, but it seems characteristic of Storage Spaces generally:

https://learn.microsoft.com/en-us/azure-stack/hci/concepts/fault-tolerance

Answer 2

Hi. Thank you for your question and reaching out. I’d be more than happy to help you with your query.

For your FIRST QUESTION:

A virtual disk in a storage context refers to a disk-like abstraction that provides access to a file or a portion of a physical storage device, but presents it to the operating system as a separate disk or volume. In the context of a storage layout, the virtual disk is created to provide a logical representation of the physical storage and allows for data to be stored in a specific manner, such as Simple, Mirror, or Parity.

A volume in the context of storage refers to a logically-defined portion of a disk, separate from the rest of the disk, which can be formatted with a file system and assigned a drive letter or mount point. In the context of the virtual disk, a volume is created on top of the virtual disk to provide a specific area for data storage, which is then managed and accessed by the operating system as a separate entity.

The creation of a column in this context likely refers to the creation of a data column in the storage layout, such as a data strip or parity block, which is used to store data in the chosen layout (Simple, Mirror, or Parity). The exact details of how this works in the background will depend on the specific storage technology and implementation being used.

For your SECOND QUESTION:

Yes, that is correct. In a two-way mirror configuration, if one disk fails, the data is still available from the other disk, which acts as a mirror. However, with an increase in the number of disks, the chances of multiple disk failures also increase. As a result, the number of simultaneous disk failures that can be tolerated by the system also increases.

In your case, with 11 disks, the number of disk failures that can be tolerated in a two-way mirror configuration would be 2. This means that if 2 disks fail, the data may become inaccessible, and the system may experience data loss.

In a three-way mirror configuration, if one disk fails, the data is still available from the other two disks, which act as mirrors. With 11 disks, the number of simultaneous disk failures that can be tolerated in a three-way mirror configuration would be 3, which provides a higher level of redundancy and reduces the risk of data loss.

If the reply was helpful, please don’t forget to upvote or accept as answer, thank you.

Answer 3

Double post

Answer 4

Thank you for your help

please, can you elaborate more about the columns and how they work in the background for the Mirror layout and parity layout, (if there is a picture it would be really good)?

As for Two-way Mirror and Three-Way-Mirror, I want to know how it calculates the increase or decrease in the number of accepted disk failures, and what is the math behind it.

understanding how the increase and decrease of disk failures are working would really save my time and help because instead of me doing a test on each disk and then adding a disk and doing a test again until I reach the number of accepted fails as below:

Two-Way:

3 Disks, max fail 1

4 Disks, max fail 1

5 Disks, max fail 1

6 Disks, max fail 2

7 Disks, max fail 2

8 Disks, max fail 3

Three-Way:

5 Disks, max fail 2

6 Disks, max fail 2

7 Disks, max fail 2

8 Disks, max fail 2

9 Disks, max fail 2

10 Disks, max fail 3

11 Disks, max fail 4

Kind Regards