11-07-2016 03:43 PM
How to add a secondary compute server and setup the load balancing using the cost alogorithm?
we already have SAS installed on a three tier architecture (Metadata, compute and mid tier) and would like to add a secondary compute server and setup the load balancing.
I found an article which says that I need to install SAS in INSTALL ONLY mode on the second compute server and then configure object spawner and then configure in Management console for the load balancing. if we do this, how will the mid tier server know about the second compute server. Any thoughts?
11-07-2016 04:03 PM
Please provide the link to the article you are referring to.
Also, what do you want to achieve by setting up a second Compute?
And what Load Balancing are we talking about, Stored Process Server?
Another instance of SAS requires a certain license, have someone at SAS (technical architect or similar) help you architect this set up?
11-07-2016 04:07 PM
Below is the link, I was referring to
we would like to distribute the load between two compute servers (workspace)
I just want to understand the process of setting up this. we will work on having a license if we want to add another instance of compute tier,
11-07-2016 04:33 PM
Kind of, but without using the grid manager. Instead would like to use the default load balancing cost algorithm. This is not as intelligent as GRID, but still load balances the sessions based on number of connections.
11-08-2016 01:46 PM
The best way to add a second compute server would be to update your existing SAS design, with a new plan XML file and do a new planned deployment. This will allow you to keep your existing deployment intact and your additional server setup consistent with whats there already.
11-08-2016 02:41 AM
11-25-2016 07:30 AM