StorageGRID Webscale Best Practices?



As per this article on AWS S3, the object key should have random prefixing to index it for better performance:


It is applicable for large number of requests and I am expecting 1000+ requests per second.


Do these best practices apply for StorageGRID Webscale too?


Do we have a best practices document available for StorageGRID Webscale too?


Re: StorageGRID Webscale Best Practices?


Anand -


StorageGrid is a passion for me, so I took notice of your post.


I found no TR docs with BCPs.

It seems our best references are the StorageGrid docs.


As an AWS fan I found your reference to the AWS doc very interesting.


I'm currently working on developement of an AltaVault course witch includes AWS and StorageGrid as cloud stores.

Any additional input you might find would be very usefull to me and other users!

I hope this response has been helpful to you.

At your service,

Eugene E. Kashpureff, Sr.
Independent NetApp Consultant
Senior NetApp Instructor, Fast Lane US
(P.S. I appreciate 'kudos' on any helpful posts.)

Re: StorageGRID Webscale Best Practices?

[ Edited ]

Hi anand1008,



StorageGRID Webscale does not require a random key prefix to improve performance. Objects are distributed for processing amongst all the nodes independent of the object key name.


You can find more information about StorageGRID Webscale and S3 here: