Forum Discussion
High Availability On-prem Gateway Cluster setup and options
Hi,
I have one on-prem gateway instance, and planning to create a high-availability cluster with two instances and "load-balancing" https://learn.microsoft.com/en-us/data-integration/gateway/service-gateway-high-availability-clusters
Please answer to all, or any of the following questions:
1) Is it possible to create an auto-scaled setup, where the second instance goes up only during high-load? Do you have an example of this kind of setup?
2) Is the automate recovery problem solved, or is the situation the same as was in 2018 https://community.fabric.microsoft.com/t5/Desktop/Automate-the-Power-BI-data-gateway-recovery/m-p/407146 ?
3) When do you prefer random load balancing between cluster instances, and when the default setting of all request to primary instance? What would be a good start, if I have two instances of different sizes?
Thank you
Antti
1) No. There is also no programmatic way to enable or disable cluster members
2) still the same. Given the instability of the gateway releases you do NOT want to automate any of this.
3) The big factor here is network performance. Whatever setup is closest to the data sources is preferred. (Which directly conflicts with BCP considerations, unfortunately). Better have multiple gateway clusters, each close to a group of sources, rather than one big cluster that is geographically dispersed.
1 Reply
- lbendlin
Super User
1) No. There is also no programmatic way to enable or disable cluster members
2) still the same. Given the instability of the gateway releases you do NOT want to automate any of this.
3) The big factor here is network performance. Whatever setup is closest to the data sources is preferred. (Which directly conflicts with BCP considerations, unfortunately). Better have multiple gateway clusters, each close to a group of sources, rather than one big cluster that is geographically dispersed.