December 13, 2017 at 6:37 am
Hi,
With SQL2016 we can run R on it's on own server or on the same server as SQL Server (different process).
The obvious benefit of the latter is avoiding data movement from db to model.
I'm thinking if we run R on it's own server in the same VLAN as SQL Server we can avoid the firewall and keep SQL Server resources targeted on the SQL workload.
Has anyone got experience of running R in these scenarios?
The workload for the SQL server is for a data warehouse with hourly data loads.
December 19, 2017 at 8:56 pm
I have not with that specific scenario. But, I think splitting them out is a great idea and it's one I would do in that instance.
For my example, I do the same for another version of SQL and it works out great. Both R and Python can run separate in their own environment with their own dedicated resources that does not hog the data warehouse. In that instance, I can directly scale up the R and Python workloads as needed without ever impacting the data warehouse and other processes that may be running.
Viewing 2 posts - 1 through 1 (of 1 total)
You must be logged in to reply to this topic. Login to reply