Building Data Dictionaries

Caitlin Hudon shares some great advice on building data dictionaries:

The best defense I’ve found against relying on an oral history is creating a written one.

Enter the data dictionary. A data dictionary is a “centralized repository of information about data such as meaning, relationships to other data, origin, usage, and format”, and provides us with a framework to store and share all of the institutional knowledge we have about our data.

As part of my role as a lead data scientist as a start-up, building a data dictionary was one of the first tasks I took on (started during my first week on the job). Learning about our data is a crucial part of onboarding for data-focused roles, and documenting that journey in the form of a data dictionary provides a useful data asset for the company (which helps to preserve institutional knowledge) and simultaneously provides a good resource for analyzing the data.

Read the whole thing.

Master Data Services No Longer Uses Silverlight

Niko Neugebauer is happy about an update to Master Data Services in SQL Server 2019:

Before we continue, let me ask you one question, have you heard about Silverlight?
Or in other words, and with a kind of evil voice “DID YOU EVER INSTALLED SILVERLIGHT ON A PRODUCTION SERVER”?.
If you have worked with MDS oh yes, you did! At least in order to check if everything is configured/upgraded correctly and nothing is broke, I will do a wild guess and claim that you did! So am I … :s

Because in order to make things work in MDS correctly, one needs this old, for a very long time deprecated framework, that is supported only in deprecated browser that is called Internet Explorer v.11, and that pain-in-the-neck framework is called Silverlight and if you dare to work with any SQL Server versions before SQL Server 2019, the picture on the left will appear in front of you at the moment you will try to explore the master data in the MDS Explorer – ensuring that unless you install a totally abandoned (and obviously unnecessary product, that represents another risk on your server) is a necessary thing. That is alone is the reason for some people would use some development VM in order to work with MDS, but that is not a good excuse to include that product in SQL Server 2016 or in SQL Server 2017.

The interface still has problems, as Niko points out, but hopefully this is the first step and not the last one.

Power BI And Many-To-Many Joins

Teo Lachev notes a big improvement to Power BI’s data modeling story:

The July 2018 preview of Power BI Desktop delivers two killer preview features that solidify the Power BI position as the best data modeling tool on the market. First, Microsoft relaxes the Power BI relationship limitations by letting you create M:M joins between two tables. Second, you can now create a composite (hybrid) data model with different storage modes, such as from an SQL Server database configured for DirectQuery and from an imported text file. Without reiterating the documentation, here are some important notes to keep in mind based on my pre-release testing.

But read the whole thing, which includes some limitations around hybrid data models.

Generating Realistic-Looking Data With Markov Chains

Phil Factor shows how to use Markov chain generation in T-SQL to generate realistic-looking country names:

How did we do this? We started with a table that took each word, added two spaces at the beginning and a |, followed by two subsequent spaces, at the end. This allowed us to map the frequency of each three-letter combination in a collection of words. Any language is made up of common combinations of characters with a few wild exceptions. For words to look right, they must follow this distribution. This distribution will change in various parts of a word, so you need all this information.

So what would happen if, instead of feeding the name of countries into the batch, we do the names of people?

My favorite name from the list was Kuwatian Samoa.

Closure Tables: Graph Data In Relational Form

Phil Factor shows how to use the concept of closure tables to represent graph-style data in a relational database:

Closure tables are plain ordinary relational tables that are designed to work easily with relational operations. It is true that useful extensions are provided for SQL Server to deal with hierarchies. The HIERARCHYID data type and the common language runtime (CLR) SqlHierarchyId class are provided to support the Path Enumeration method of representing hierarchies and are intended to make tree structures represented by self-referencing tables more efficient, but they are likely to be appropriate for some but not all the practical real-life hierarchies or directories. As well as path enumerations, there are also the well-known design patterns of Nested Sets and Adjacency Lists. In this article, we’ll concentrate on closure tables.

A directed acyclic graph (DAG) is a more general version of a closure table. You can use a closure table for a tree structure where there is only one trunk, because a branch or leaf can only have one trunk. We just have a table that has the nodes (e.g. staff member or directory ‘folder’) and edges (the relationships). We are representing an acyclic (no loops allowed) connected graph where the edges must all be unique, and where there is reflexive closure. (each node has an edge pointing to itself)

Take the time to read this one carefully, as I think this model is applicable much more often than it’d appear at first blush.

SQL Server’s Referential Integrity Operator

Joe Obbish explains the purpose of the referential integrity operator in SQL Server 2016:

What would happen if a parent table was referenced by hundreds of child tables, such as for a date dimension table? Deleting or updating a row in the parent table would create a query plan with at least one join per incoming foreign key reference. Creating a query plan for that statement is equivalent to creating a query plan for a query containing hundreds or even thousands of joins. That query plan could take a long time to compile or could even time out. For example, I created a simple query with 2500 joins and it still hadn’t finished compiling after 15 minutes. That’s why I assume a table is limited to 253 incoming foreign key references in SQL Server 2014.

That restriction won’t be hit often but could be pretty inconvenient to work around. The referential integrity operator introduced with compatibility level 130 raises the limit from 253 to 10000. All of the joins are collapsed into a single operator which can reduce compile time and avoid errors.

There’s some really good information in this post, and Joe has mixed feelings on the concept.

Thinking About The Data Lake

James Serra explains at a high level what the data lake metaphor is and how it works:

The data lake introduces a new data analysis paradigm shift:

OLD WAY: Structure -> Ingest -> Analyze

NEW WAY: Ingest -> Analyze -> Structure

This allows you to avoid a lot of up-front work before you are able to analyze data.  With the old way, you have to know the questions to ask.  The new way supports situations when you don’t know the questions to ask.

This solves the two biggest reasons why many EDW projects fail:

  • Too much time spent modeling when you don’t know all of the questions your data needs to answer

  • Wasted time spent on ETL where the net effect is a star schema that doesn’t actually show value

There are some good details here.  My addition would be to reiterate the importance of a good data governance policy.

Re-Shaping Data Flows

Maneesh Varshney explains some methods to trim the fat out of analytical data flows:

Big data comes in a variety of shapes. The Extract-Transform-Load (ETL) workflows are more or less stripe-shaped (left panel in the figure above) and produce an output of a similar size to the input. Reporting workflows are funnel-shaped (middle panel in the figure above) and progressively reduce the data size by filtering and aggregating.

However, a wide class of problems in analytics, relevance, and graph processing have a rather curious shape of widening in the middle before slimming down (right panel in the figure above). It gets worse before it gets better.

In this article, we take a deeper dive into this exploding middle shape: understanding why it happens, why it’s a problem, and what can we do about it. We share our experiences of real-life workflows from a spectrum of fields, including Analytics (A/B experimentation), Relevance (user-item feature scoring), and Graph (second degree network/friends-of-friends).

The examples relate directly to Hadoop, but are applicable in other data platforms as well.

Trimming The Fat From Power BI Data Models

Philip Seamark shows how things like getting rid of ID columns can reduce a Power BI data model’s size significantly:

Once saved, the Power BI file size was 289MB!  Is this good for 10 million rows?  It’s certainly better than the 360MB CSV file but not by much.  Certainly not close to the 10:1 compression claimed to be achievable using the SSAS Tabular engine used by Power BI.

I think we can do better than that….

Read on to see the specific optimizations, turning this from a 289 MB data model into a 9 MB data model.

Dimensional Design Tips

Koen Verbeeck provides some helpful hints when designing dimensions in SQL Server Analysis Services Multidimensional models:

Although traditional dimension modeling – as explained by Ralph Kimball – tries to avoid snowflaking, it might help the processing of larger dimensions. For example, suppose you have a large customer dimension with over 10 million members. One attribute is the customer country. Realistically, there should only be a bit over 200 countries, maximum. When SSAS processes the dimension, it sends SELECT DISTINCT commands to SQL Server. Such a query on top of a large dimension might take some time. However, if you would snowflake (aka normalize) the country attribute into another dimension, the SELECT DISTINCT will run much faster. Here, you need to trade-off performance against the simplicity of your design.

There are several good tips here.

Categories

November 2018
MTWTFSS
« Oct  
 1234
567891011
12131415161718
19202122232425
2627282930