MakeLine() + Transparency + Mark Hiding = Oh My!

Since I arrived at Snowflake, I’ve been using a dataset from the NYC Citibike program as my demo data. It’s about 61M bike trip records with lat/lon data for the starting and ending bike stations, and I’ve been showing it using the following dashboard where the starting and ending locations are in different maps…

What I really wanted to do though was to make a map that allowed me to select a starting station and show where passengers went. Something like this…

Well, with the latest 2019.2 beta, I did (as you can see). It’s running over my Snowflake demo database – naturally – and uses the following calculation to create the line:

TripLine = 

  MAKEPOINT([Start Station Latitude], [Start Station Longitude]),
  MAKEPOINT([End Station Latitude], [End Station Longitude])

Oh, how long I have waited for thee, sweet MakeLine() and MakePoint()…

The reason this new feature helps so much is that the starting and ending location data is in the same record. In earlier versions of Tableau I would have had to pivot the data to split the record into two so I could use a line marker and the path shelf. This would have performance implications as I would be effectively doubling the amount of data we need to read and process.

So MakeLine() gives me the line between the start and end points, which I display in a sheet:

I also want the point map for the starting locations (so I can select them) – here’s that sheet:

As you can see, both of these maps are set up with no background. Thanks to this blog post ( that showed me the trick for layering maps and getting the transparency to work. The secret is to set the sheet shading to be “none”:

Once I did this, I was able to layer them together using floating placement:

To get them to line up, I used the search/zoom feature of the map to set the viewport on each map to the same region:

Now to make the two maps work together… I set up a filter action from the point map to the line map to only show the lines for the selected starting station:

I also need a filter to control the Num Trips panel in the top right – this needs to be a separate filter as the logic is reversed to the lines. I.e. when I clear the selection, I want all the trips to be shown:

This gives me the interactive behaviour between the three elements (points, lines, trip count) but the problem is that the points layer is obfuscating the line layer (because it’s on top):

We need to hide the unselected points, and our good old “hide the mark using an invisible shape” trick comes into play. We define a set that will hold the selected marks, and we use the set to control the mark shape – using a filled circle for IN the set and an invisible 10x10px PNGcustom shape for OUT of the set:

We then use an Edit Set action to update the set values to the selection:

The result – the selected points are shown as circles while the unselected points disappear, allowing us to see the lines unobstructed. We get all the points back by clearing the selection:

The final point is that we need the base map for spatial context. I initially had this in the lines map but the problem was that it disappeared when there were no lines to show – i.e. the case when no points are selected.

To overcome this, I simply added a 3rd map sheet with a visible base map but with a single mark that was invisible. This meant that the map would be visible in all states of the upper layers. Again, transparency allows this to show through:

The final result is a map with both points and lines that dynamically updates based on the selection. The key elements were:

  • MakeLine() to create the line mark between the starting and ending stations, even though there is no spatial data elements in the underlying database – just raw lat/lon FLOAT fields. Also, we don’t need to pivot the data to make starting and ending records;
  • Sheet transparency to allow me to layer the maps together;
  • That good old dynamic mark visibility trick using an invisible custom shape; and
  • Edit Set actions to control the visibility of the point marks.

I’ve uploaded a version of the workbook with a reduced set of data extracted so you can pull it apart.


Posted in Uncategorized | 1 Comment

Tableau and Snowflake Boolean fields

Hello, gentle reader. It’s been a while. I hope you are well. 🙂

Today, I was helping a colleague with a problem that stems from the fact that the current version of the Snowflake driver in Tableau doesn’t correctly support Boolean fields and therefore returns data from Snowflake BOOLEAN fields as a 1/0 integer value. This means that the field cannot be directly used in a Boolean expression like:


where A and B are both themselves Booleans. So I thought I’d do some digging to understand what was happening under the covers.

For this particular problem, I came up with four different solutions… combinations of using RAWSQL_BOOL to force the result to be interpreted as a Boolean, and simple integer comparisons that return a Boolean result:

  • Create individual Boolean fields for A and B using RAWSQL_BOOL(), then create a third field that ANDs them together.
    • Formula:
      • [A_BOOL] = RAWSQL_BOOL(“A”)
        [B_BOOL] = RAWSQL_BOOL(“B”)
        [A&B] = [A_BOOL] and [B_BOOL]
    • Produced the following SQL:
  • Create a single field that embeds the AND into the RAWSQL_BOOL function. Less flexible than option 1 but maybe better performing?
    • Formula:
      • [A&B] = RAWSQL_BOOL(“A AND B”)
    • Produced the following SQL:
  • Create individual Boolean fields for A and B using an integer comparison, then creating a third field that ANDs them together. My thought here was that integer queries are generally very fast so this might be faster than the RAWSQL approach…
    • Formula:
      • [A_INT] = [A] = 1
        [B_INT] = [B] = 1
        [A&B] = [A_INT] and [B_INT]
    • Produced the following SQL:
      • (CASE WHEN ((“BOOL_TEST”.”A” = 1) AND (“BOOL_TEST”.”B” = 1)) THEN 1 WHEN NOT ((“BOOL_TEST”.”A” = 1) AND (“BOOL_TEST”.”B” = 1)) THEN 0 ELSE NULL END)
  • Create a single field that does an integer comparison for each field and ANDs the results. To be honest, I didn’t expect much difference between this approach and option 3…
    • Formula:
      • [A&B] = [A]=1 AND [B]=1
    • Produced the following SQL:
      • (CASE WHEN ((“BOOL_TEST”.”A” = 1) AND (“BOOL_TEST”.”B” = 1)) THEN 1 WHEN NOT ((“BOOL_TEST”.”A” = 1) AND (“BOOL_TEST”.”B” = 1)) THEN 0 ELSE NULL END)

I put the four examples above into a Tableau worksheet and the following query was run in Snowflake:

SELECT (CASE WHEN (A AND B) THEN 1 WHEN NOT (A AND B) THEN 0 ELSE NULL END) AS “Calculation_ 1822761623506259971”,
(CASE WHEN ((A) AND (B)) THEN 1 WHEN NOT ((A) AND (B)) THEN 0 ELSE NULL END) AS “Calculation_ 1822761623506604036”,
(CASE WHEN ((“BOOL_TEST”.”A” = 1) AND (“BOOL_TEST”.”B” = 1)) THEN 1 WHEN NOT ((“BOOL_TEST”.”A” = 1) AND (“BOOL_TEST”.”B” = 1)) THEN 0 ELSE NULL END) AS “Calculation_ 1822761623507091461”,
(CASE WHEN ((“BOOL_TEST”.”A” = 1) AND (“BOOL_TEST”.”B” = 1)) THEN 1 WHEN NOT ((“BOOL_TEST”.”A” = 1) AND (“BOOL_TEST”.”B” = 1)) THEN 0 ELSE NULL END) AS “Calculation_ 1822761623510069255”,
SUM(“BOOL_TEST”.”C”) AS “sum:C:ok”

Looking at the query profile, the two Boolean approaches produced the same expression in the query profile:


As did the two integer approaches:


Finally, I timed a series of tests against an admittedly small data volume (~100K rows) and the integer logic seemed to come out slightly faster on average – 140ms vs. 166ms over 12 iterations. I’m not convinced that the result is statistically valid, but read into it what you will.

If you want to have a play around with this yourself, here’s the create table statement:


The data for the table and the Tableau workbook can be found in my Dropbox.

Overall, I’m not sure if that helps but it certainly explains what was going on under the covers. It also gives some workaround options until the problem is fixed by Tableau.


The objective of this piece was to discuss how you could use Boolean values in Boolean logic formulae, however, there is one additional piece worth talking about… how to get the field to display as a Boolean true/false value? While the approaches above create fields that are true Booleans in Tableau, there is another approach that doesn’t require creating a calculation, and that is using an alias on the field. You can apply an alias of “true” to the numeric value of “1” and “false” to the numeric value of “0”. This would then display the alias value but the field is still a numeric for the purposes of calculations and such.

I just thought it was worth noting this additional option.

Posted in Uncategorized | Leave a comment

Snowflake multi-cluster warehouse vs. TabJolt load

One of Snowflake’s key features is something called multi-cluster warehouses.

By default, a virtual warehouse consists of a single cluster of servers that determines the total resources available to the warehouse for executing queries. As queries are submitted to a warehouse, the warehouse allocates resources to each query and begins executing the queries. If sufficient resources are not available to execute all the queries submitted to the warehouse, Snowflake queues the additional queries until the necessary resources become available.

With multi-cluster warehouses, Snowflake supports allocating a larger pool of resources to each warehouse. As the number of concurrent user sessions and/or queries for the warehouse increases, and queries start to queue due to insufficient resources, Snowflake automatically starts additional clusters, up to the maximum number defined for the warehouse.


Similarly, as the load on the warehouse decreases, Snowflake automatically shuts down clusters to reduce the number of running servers and, correspondingly, the number of credits used by the warehouse.

As you can imagine, this capability is very useful to maintain a consistent response time for users in BI and reporting scenarios where varying user loads are common. Imagine that you normally have 5 concurrent users on your system but on Monday mornings, you have a spike of 20 concurrent users, all wanting to view reports for the weekend’s business. A multi-cluster warehouse can ensure that users during the peak load experience the same response times (from the DB that is… saturation of the BI server is a separate factor) as users during non-peak periods.

To show this capability in action, I created a dashboard that queries against about 57M records of Citibike data, hosted in a Snowflake DB. I published this dashboard to Tableau Server and then used TabJolt to simulate a) a single user baseline, and then b) a peak period with multiple concurrent users. Note that caching was turned off both in Tableau Server and in the Snowflake data source to ensure that all queries were actually run in the DB.

You can see this experiment and the results in the following video. Apologies for just providing a download link, but I tried posting this to Vimeo but the video quality was terrible. That’s most likely my fault, but right now this is an easier way to share:


Posted in Uncategorized | Leave a comment

Snowflake and Tableau – in action!

A few days ago I was invited to present Snowflake as part of a webinar run with our partner BockCorp. Check out the recording of the session here, which includes an overview of the Snowflake architecture as well as a demo showing all the cool capabilities like instant elasticity, semi-structured data support, data sharing and more:

Oh – if you want to jump straight to the demo, it starts around the 29min point. Go ahead, I won’t be offended. 🙂


Posted in Uncategorized | Leave a comment

Best Practices for Using Tableau with Snowflake

Update 18 July 2018:
The whitepaper has now been released as an official Snowflake whitepaper so I’ve updated the download link to point to the document on the Snowflake website. The new whitepaper is much prettier and has been through editing to clean up all my bad writing habits. Thanks to Vincent Morello and Marta Bright in our content marketing team for all their help in making this happen.

As announced in my last post, since joining Snowflake I’ve been working on a whitepaper that provides best practice guidance for using Tableau with our built-for-the-cloud data warehouse.

Well, I’m pleased to report that it’s done. Or at least, done enough to release. You can download it from here:

I hope you find it useful, and please let me know if you have any feedback or corrections.

Posted in Uncategorized | 2 Comments

Tableau and Snowflake

Happy New Year everyone!

I’ve been a bit quiet of late. Probably to be expected, what with getting my head around all the new stuff here at Snowflake. Also, properly relaxing over Christmas and summer requires a degree of focus (ah, the joy of the southern hemisphere!). But I’ve not been completely idle. Over the past few weeks I’ve been steadily working on a new whitepaper:


Here’s an overview of the document scope…

  • Introduction
  • What is Tableau?
  • What is Snowflake?
  • What you DON’T have to worry about with Snowflake
  • Creating Efficient Tableau Workbooks
  • Connecting to Snowflake
  • Working with Semi-Structured Data
  • Working with Snowflake Time Travel
  • Working with Snowflake Data Sharing
  • Implementing Role-Based Security
  • Using Custom Aggregations
  • Scaling Snowflake Warehouses
  • Caching
  • Other Performance Considerations
  • Measuring Performance

Of course, it’s turning out to be a lengthy read – it seems I know no other way. 🙂 But believe me, a lot of that is screenshots and SQL. The document is being reviewed at the moment, but I plan to break it into consumable chunks and release material as posts over the next couple of weeks. Maybe here, maybe on the Snowflake or Tableau blogs.

So, keep your eyes peeled…

Posted in Uncategorized | 4 Comments

Endings and Beginnings

Well it was a bittersweet day on Friday. After 6+ years at Tableau I have decided that it’s time for a new challenge. Tomorrow I start my first day at Snowflake Computing, a company that is revolutionising the cloud analytic database market.


I’m going to continue to blog here and given I still have a deep love for Tableau, some (many) of my posts will continue to be about it and data visualisation in general. However, I’ll also be posting about Snowflake and interesting things I’m learning as I settle in to my new role. Given that a primary use case for Snowflake is BI and analytics, the two topics should be quite complementary.

Thanks for your support and questions over the past few years and I hope you continue to find my ramblings informative.

Posted in Uncategorized | 8 Comments