Data freshness policies in Tableau Server & Online : New in Tableau 2021.3
Tableau 2021.3 finally lets you dictate exactly how fresh your live-connection data needs to be.
- Data freshness policies only apply to live connections; embedded extracts don't expose the option.
- Access the setting via the eye icon on a workbook's details, not the three-dot menu.
- The site default is 12 hours, but you can set freshness from minutes up to weeks.
- You can schedule a specific time and time zone for data to refresh, similar to extract schedules.
- Caching happens at multiple points (Tableau and the database), so a 'live' request still may return cached data from the database itself.
- You can't set multiple refresh times per day; use a recurring interval like every three hours instead.
0:00Hey Tim here in Tableau 21.3 Tableau have
0:02added this ability to essentially dictate
0:05the data
0:06freshness policy for live connection in a
0:08particular workbook. Let's take a look and
0:11see
0:11how that works. Now, there is one condition
0:13for this is that you have to be using a
0:14live connection.
0:15This doesn't work on a extracted workbook.
0:18Let's let me just show you this. You can
0:20see here this
0:20doesn't have a live connection to a data
0:22source. If you go to data sources here, you
0:24can see that
0:25these are just two extracts embedded inside
0:27of the workbook. If I go back to the views
0:29option,
0:30this is sort of tricky because you might
0:31think you hit on these three dots to see
0:33this option.
0:34If you click on that, you just get this
0:35drop down. What you actually need to do is
0:37click on this
0:38little eye. This is where you get the
0:40details of this particular workbook. And
0:42you can see here
0:43that I don't get any options for data fresh
0:46ness. It should normally be here and it's
0:48not because
0:48this is essentially an embedded data source
0:50and it's extracted. It's not a live
0:52connection,
0:53so none of the caching really matters here.
0:55Okay, let's go back to my home page. And
0:58what you'll see
0:58is I actually created a workbook literally
1:01two minutes ago. And we're going to use
1:03that. Now
1:04what this is actually doing, let's go ahead
1:06to this workbook. It's a really crap work
1:08book. I've
1:08just created it as a simple demo. And it
1:11essentially shows the view counts on my
1:14videos,
1:14spread out across a bunch of days. Now, the
1:16secret here is that obviously it's messy,
1:18so you can't make it out and that the video
1:20IDs are over here, because no one watches
1:22my videos.
1:22True story. But nonetheless, if we go back
1:25out to the workbook up here, you'll see
1:28that this
1:29actually is using a data source. It's a
1:31live connection and it's connecting to Snow
1:33flake
1:33and it's using a table in my Snowflake
1:35instance to do this. So this is a live
1:37connection, this
1:38should work. So if I go over to the top
1:41here and hit the eye icon, you'll see that
1:43essentially I
1:44now get this data freshness policy area.
1:47And it's actually set to the site default
1:49because I haven't
1:50dictated one, it's got the site default of
1:5212 hours. So essentially Tableau is going
1:55to make
1:55sure the data is as fresh as it possibly
1:58can be over the last 12 hours essentially.
2:01And it's sort
2:02of hard to explain caching in Tableau
2:04Server and online because there's so many
2:07different points
2:08where caching can happen. Caching can
2:10happen in Tableau, caching can happen in
2:12the database. So
2:14even though this says 12 hours, the
2:15database might be doing something
2:16completely different.
2:18So it's really hard to diagnose caching
2:19sometimes, but what this does do is it
2:22gives a level of
2:23certainty about what Tableau is trying to
2:25do, which has always been sort of difficult
2:28with
2:28different points of caching inside of Table
2:30au itself. So you can see here that I've got
2:32the
2:33site default of 12 hours. Obviously if you
2:35hit learn more, it opens up a new tab. This
2:37goes to
2:37the documentation on this feature. So by
2:40all means, absolutely check this out. But
2:42if I go back to
2:42this feature, you'll see that I have this
2:44setting. Now I can always say that this is
2:46live. Tableau
2:47will always get the latest data. This is
2:50important because for some workbooks and
2:52for some use cases,
2:53this is literally the only way to make
2:55absolutely certain that you're looking at
2:57the latest data.
2:58Now this just means that Tableau will send
3:01a request to the database. It doesn't mean
3:04the
3:04database itself will actually send back
3:06something that's not cached. So that's
3:09something to sort of
3:10watch out for. But you can also change this
3:12value of 12. So if I go down here to the
3:14next option,
3:15ensure the data is fresh, I can change this
3:17value every six hours, five hours. I can
3:20even go all the
3:21way up to weeks. If I've got a live
3:22connection that barely changes, I can say
3:24to Tableau, hey,
3:26keep this cache for up to a month. I don't
3:28know when you'd ever use that. But you can
3:30also set
3:30it as low as minutes. So you can say, hey,
3:33every 10 minutes, can you please make sure
3:35that this
3:36is up to date? Okay, so this is a really
3:38nice sort of precise control that we now
3:41have. And you can
3:42set this to a really nice setup. Now, I can
3:45also specify a time this is so powerful. So
3:49if you look
3:49at this, I can actually specify that listen
3:52, at 945, when people log in, after going
3:56through,
3:57I don't know, having their coffee or
3:58whatever, I look at the stat, I can say to
4:00Tableau, hey,
4:01at nine o'clock, every single day, every
4:04single week, whatever you want to do, you
4:06can basically
4:07set this schedule exactly like you can
4:09almost set extract schedules. They look
4:12Monday, Thursday,
4:13Wednesday, Thursday, and Friday at nine o'
4:16clock, using the American time zone here,
4:18because this
4:19server is in America, but I can choose
4:21whatever timezone I want. Can you make sure
4:23that this
4:24data source is using live fresh data,
4:27please. And that will make sure that it's
4:30always pulling that
4:31in from that time onwards. So if it gets to
4:33859, it won't. And then at nine o'clock,
4:36boom, every
4:36request at that point onwards will be fresh
4:38. And so the caching will start to pick up
4:40there. And so
4:41this is a nice way to sort of make this
4:44work. Now, the frustrating thing about this
4:46setting is that
4:47you can't set multiple points in the day.
4:49So what you have to do if you want to do
4:51this is really
4:52go back to this option here, and say look
4:54every three hours. And then generally, this
4:57should catch
4:58most of the requirements that you want, I
5:01would like to be able to just type in the
5:03space,
5:04it doesn't seem to be working. It's just a
5:06sort of up and down toggle at the moment.
5:08But that's
5:08pretty much it. That's sort of the feature
5:10in a nutshell, you can use this to really
5:12set how fresh
5:13your data is going to be across the whole
5:15entire tableau product. So that's really
5:16cool to see.
5:17And that's really cool to have. And yeah,
5:19if you enjoyed this video, thanks for
5:20watching. Be sure
5:21to check out all the other videos on Table
5:23au Tim calm, where I go through all these
5:25features for
5:26this release and the past released as well.
5:28And we cover lots of other great content as
5:30well. So
5:30be sure to check that out. Let me know in
5:32the comments what you'd like to see more of
5:34and I'll
5:34catch you in the next video.
5:35Transcribed by https://otter.ai
Control the age of the data shown to workbook viewers by setting a freshness policy that will not load any cached data older than the set policy. Easily choose from multiple policy options to ensure that workbooks with live data sources never have unexpectedly stale data.