tryllz

Understanding Disk Array Availability..

Recommended Posts

Hi,

I'm new to this thus I'm trying to understand about calculating Disk Array Availability (DAA).

What I have understood so far is that availability is always 1, and Failure = 1÷MTBF.

Could someone explain in simple English how to calculate DAA.

Assuming I have 1TB and the time is 5 years, and the disk's MTBF is 1.6 millions hours (got this from Dell's website for a 15K RPM disk), I need to know the availability of this disk in 5 years, how do I calculate, what is the formula, I'm seeing , MTTl, MTTR, and MTTDL and getting confused.

Another confusion is, is disk array availability associated only with RAID.

Appreciate the assistance.

Currently I'm using this as reference : http://www.ecs.umass.edu/ece/koren/architecture/Raid/reliability.html

Also in the reference it says double disk failure, does the double disk mean any 2 disks or what, because once we find the availability of 1 disk would multiplying that into 2 NOT give us the double disk failure..

If anyone knows of any other good reference that explains in simple English.

Thank You

Share this post


Link to post
Share on other sites

The link you have refers to disk arrays, which in general definition are always composed of multiple disks (hence the name, disk array) whereas you appear to be referring to a single disk. With a single disk there's no availability calculation to be done beyond MTBF.

A more useful metric for you to worry about might be AFR, annualized failure rate, for your particular model of disk. Check the specs for that to get what should hopefully be an accurate spec (or speculation) on the odds of your single disk failing over the course of a year. Then you can start looking up formulas that calculate failure rates out to five years.

 

And if it helps any, availability is most certainly not 1 all the time, although that may be the desired goal!

Share this post


Link to post
Share on other sites
1 hour ago, continuum said:

The link you have refers to disk arrays, which in general definition are always composed of multiple disks (hence the name, disk array) whereas you appear to be referring to a single disk. With a single disk there's no availability calculation to be done beyond MTBF.

A more useful metric for you to worry about might be AFR, annualized failure rate, for your particular model of disk. Check the specs for that to get what should hopefully be an accurate spec (or speculation) on the odds of your single disk failing over the course of a year. Then you can start looking up formulas that calculate failure rates out to five years.

 

And if it helps any, availability is most certainly not 1 all the time, although that may be the desired goal!

Thanks for the reply from what I understand is that calculating availability for 1 disk is not the same as calculating for the disk array, I was under the assumption if we get value for 1 disk and multiply it by the number of disks in the disk array would get me the result for the disk array that is why I added an assumption of 1 disk.

In general is there a formula for calculating availability of a disk array.

Thank You

Share this post


Link to post
Share on other sites

Disk arrays come in multiple possible configurations. You would need to determine how many disks you will be using in the array and the configuration you are using.

Some configurations have zero tolerance for any failures and hence are actually less (potentially significantly so) reliable than a single disk, others can handle a single disk failure, others two disk failures, others ... well let's just say it depends where in the array the second (or third, or fourth, etc.) disk fails.

You would need to look up the formulas for each array configuration you are looking at.

 

I did break this all down (or at least find the links) about five years ago, but unfortunately the sources I use escape me at the moment. The calculator linked in the link you posted above is useful enough to have some utility if you want a purely mathematical calculation for the two specific array configurations it covers.

 

Quote

MTTl, MTTR, and MTTDL and getting confused.

It's actually explained in the link you have, but as you can see, disk array availability is rather complicated. Calculating annualized failure rate (for a single disk) or mean time to failure is more simple and is sufficient for most. If you are still intent on calculating array availability then you may want to ask why you are doing so, and understanding the mechanisms that affect it.

aka find some links that explain things with better examples than the glossary in the page you linked?

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now