Solved

Need to understand the available Storage when one node fails when a cluster is RF2

  • 23 September 2021
  • 1 reply
  • 81 views

  1. Hello,

    I have a 3 node cluster with RF 2 configured. When All the nodes are up and running I can see Total Cluster resources as below on the Prism Element.
  • CPU - 388 Ghz
  • RAM - 4.5 TB ( 1.5 TB Per Node)
  • Storage - 32.9 TB 

 

Let’s say If 1 node fails, Will my Available Storage size reduce? Please Help me.

 

  1. Also, I don’t understand one thing, I have Total 42 SSD of 1.92 TB = 80.64 TB, But in reality with RF2 why am I getting only 32.9 TB (without any data efficiency options enabled). I should Ideally get 40 TB of usable space. I checked on the storage sizer portal and it said I should get 63 TB. Please guide guys, what am I missing.

     

 

icon

Best answer by mikkisse 23 September 2021, 15:36

View original

This topic has been closed for comments

1 reply

Userlevel 3
Badge +1

Hello!

I think this article can answer to all your questions:

Recommended guidelines for maximum storage utilization on a cluster

Also you operating with TB (Terabytes), but Nutanix shows you data in TiB (Tebibyte). Here is a calculator - https://www.gbmb.org/tb-to-tib

Also check 1st and 2nd disks on your nodes, it should be less than other similar disks in size, because part of it used for CVM needs. (Hardware - Diagram)

 

In your case you should not use more that 20TiB of storage in RF2, then you can tolerate to one node failure situation. 

Sum of all your disks capacity on one node = physical node capacity;

Sum of all your nodes = physical capacity of your cluster;

logical capacity = physical / 2 in your case;

resilient capacity = (logical capacity - one node logical size) * 0,95. (this is my assumption).