Capacity Management - On -premises Gateway Server

[Anonymous]

Current scenario:

• We are using Pro licence and are using one On-premises gateway server ( Windows 2016 , Specification 4vCPU-32GB RAM-250GB) to connect to On-premises data sources.
• We are encountering performance issues,time-out errors and the data refresh failures due to high CPU and memory usage.

Can anyone please help with the below :

1. Can the gateway server be further scaled up, if yes, what is the CAP limit for RAM, CPU, etc?

2. Is there any detailed document that can guide us to determine the following:

1) How to utilise the Usage stats to determine if scale-up or scale out is the best option to overcome the issue?

2) If scale-up is the option, how to perform the capacity planning calculation based on current consumption? Is there a standard baseline that we can refer to?

3) If scale-out is the option, how to decide if one cluster with multiple node or multiple cluster with one or 2 nodes would be more effeicient and cost-effective and the use-cases for each of thse option?

4) If scale-out is the option, how to calculate how many number of nodes should be added in a cluster based on current consumption? Is there a standard baseline that we can refer to?

[Anonymous]

Hi @Anonymous ,

Is this issue solved now? If you have any questions, please feel free to share with us.

And don't forget to accept the answers if they helped. Your action would be helpful to other users who encounter the same issue and read this thread. 🙂

Thanks for your understanding.

Best regards,

Lucy Chen

[lbendlin]

Your memory looks good (make sure it is actually used!)  but the vcores need to be increased - ideally to 16.

For the disk space - place spooler and reports files onto a different spindle, and enable StreamBeforeRequestCompletes

What's the NIC speed? What is the typical workload - more mashups or more direct queries?

And finally - how many of these VMs do you have in your Gateway Cluster?