It sounds like your organization has already made a decision however if that isn't the case, I suggest contacting your Tableau Account manager and starting a free trial of Tableau Server. The features of Tableau Server are outlined here Tableau Server | Tableau Software and the setup and deployment are somewhat dependent on your data sources, size of data and the way and number of users that will be looking at the visualizations.
A detailed review of your organizations needs is best done by a member of that team.
If you do not have a Tableau Account manager contact yet, just click on that 'Contact Us' link at the bottom of the page.
I hope that helps
Thanks for replying. we have Tableau Server but it may be the case that we are not using it to it's full potential.
I'll contact our rep tomorrow.
Are there any resources for how to use/organise server for better end user experience?
Hey Edward -
Do you happen to have links to the material that says "embedded stuff" doesn't get cached. That's not correct as far as I know - so I'd like to see the original source material.
This typically isn't a big deal or a hard thing to do. For example, here's something that Fannie Mae did with Tableau: Freddie Mac MiMi -- Metro
Thanks for this.
Here are the two threads that alerted me to the issue:
Embedding dashboards in our work SharePoint is the ideal option for us as its straightforward to do, looks pretty good, and colleagues already use the sharepoint regularly.
I did some tests earlier today, putting the same dashboard both on the sharepoint and on the server and could see any noticeable difference in performance between the two which was interesting and reassuring to see.
2 of 2 people found this helpful
I looked at this a little bit more closely, and it worked as expected for me (which pretty much matches your results);
I executed a dashboard (which takes about 20 seconds to render - fires a bunch of heavy Redshift Queries) 5 times:
- First time: Interactively in the portal.
- Third time: In an HTML page using an IFRAME
- Fourth time: In an HTML page using an IFRAME AND using the embed=y parameter
- Fifth time: in an HTML page using an IFRAME AND the embed=y AND refresh=y parameters
As you can see, rendering is not hitting cache during executions 1 and 5. We hit the cache for executions 2, 3, 4.
Here are the (parsed) logs sitting in postgres. They tell the story pretty well.
Beginning at 16:53, you can see 5 query id's in a row (0-4) which represent different queries being executed against Redshift. They all share the same vizql session id, so you know they are coming "together"
Then, at 16:55, 16:57.04- .05 and 16:57.17, you see some literal caching going on for 3 other vizqlsessions
Finally, at 16:58, more "real" queries get executed...
Thanks for your time on this Russell.
It's great to get this confirmed and with such detailed backup.