The Tessitura upgrade instructions for v15 suggest using two separate servers; Analytics Database Server and Analytics Cube Server. We are considering combining the two functions onto one server.
Are there any considerations regarding CPU, memory or I/O that would warn against combining the two?
Thanks,
Mark
Hi Mark,
The recommendations around having a separate SQL database server for the Analytics data warehouse, separate from the Analytics Web/Cube Applications server, stems from understanding that a number of organizations have T-Stats installed and would like to avoid additional SQL licensing. In that scenario, adding the Analytics data warehouse to that same T-Stats SQL instance, and then putting the Analytics Web/Cube Applications on a separate server addresses that concern. Removing T-Stats from the mix, then putting all three Analytics components on a single server is the recommended, basic configuration. This recommendation assumes that the server is dedicated to the 3 Analytics components.
Best,Chris
Chris Wallingford Director, Business Intelligence Tessitura Network Office +1 888.643.5778 ext 553 chris.wallingford@tessituranetwork.com Tessitura Learning & Community Conference July 28-August 1, 2019 #TLCC2019
Chris, while I have your attention, one more question. The docs requirements are for (up to) Server 2016 and SQL 2016. We're planning on using Server 2019 and SQL 2017. We assume that will be fine but would like your opinion on that.
We have been successful running SQL 2017 in SQL 2016 Compatibility mode. Testing still needs to be conducted for SQL 2017 run without being in compatibility mode, so I cannot comment on that at this time. As for Windows Server 2019, we are just starting to build an environment for testing and therefore have cannot give full consent on how it will work with Tessitura and Analytics.
Karen
Thank you Karen.