Skip to main content

could someone explain me how the below options are working ,

 

if i have 6 nodes for a storage pool , what if i have more than 6 nodes , 

what is the disperse factor mentioned here 

 

Hi @Ajal 

For HyperScaleX, we have metadata nodes which are generally the first three nodes for deployment. Irrespective of how big the cluster is, only one node failure can be tolerated.

You can run the command /usr/local/hedvig/scripts/showmembers.exp to find the metadata nodes. It will show port 7010 for the nodes. Nodes that show only 7000 are not running metdata service.

hedvigduro> showmembers
xxx.prod:7000
xxx.prod:7010
yyy.prod:7010
yyy.prod:7000
zzz.prod:7010
zzz.prod:7000

 

The EC code will be 4+2, even if you have more than 6 nodes.

 


@R Anwar  then what is the point of giving option to select standard medium or high,

 

which one to select ? 


Hi @Ajal 

I reconfirmed and 3 nodes with have EC 4+2 and 6 nodes and above will have resiliency of 8+4 giving higher tolerance level.


@R Anwar thanks for the update

 

could you tell me what is meant by disperse factor here also if i choose high the usable capacity will be less?


Hello @Ajal 

Please refer to get an understanding on the terminologies : https://documentation.commvault.com/2023e/expert/9092_hyperscale_reference_architecture_terminology.html

Best,

Rajiv Singal


Reply