Tony Hillman there is slight scope increase in the requirements for this ticket. Along with documenting the new limit, we also want to clearly state certain guidelines when using high number of buckets. The guidelines are as follows:

• Sizing has to account for raw CPU overhead of high number of buckets. It is recommended to set aside 0.4 core/bucket for rebalance to work smoothly. This overhead does not account for any front-end workload which must be sized separately.
• Large number of indexes per bucket may cause issues when there are high number of buckets. It is recommended to stay below 10 indexes per bucket.
• There are known issues with Eventing service and large number of buckets.
• System limits for file descriptors and TCP-IP connections may be exceeded and should be adjusted.
• Monitoring is recommended for CPU usage and System Limits

Ian McCloy FYI.

Tony Hillman is this done? We need this fixed for the Beta.

amarantha [X]?

Adding "must-fix" for this ticket. amarantha [X] fyi.

To simplify, we can state the following guidelines only:

• Sizing has to account for raw CPU overhead of high number of buckets. It is recommended to set aside 0.4 core/bucket for rebalance to work smoothly. This overhead does not account for any front-end workload which must be sized separately.
• Large number of indexes per bucket may cause issues when there are high number of buckets. 
• Monitoring is recommended for CPU usage and System Limits

Amarantha Kulkarni / Tony Hillman do you have an ETA for getting this change in?

Created a new ticket to track the addition of sizing guidelines - DOC-8454 (ETA 4/23/2021)

Will resolve this ticket as the initial request to update the bucket limit to 30 has been completed.

The initial task to document increased bucket limit has been completed. 

Additional note to add sizing guidelines is being tracked in a separate ticket: DOC-8454 - Add sizing information to Server docs