SSIS Series | SQL Freelancer Blog

SSIS Series: How to use Conditional Split

By bradyupton in Business Intelligence, Integration Services (SSIS), SSIS Series January 30, 2023 0 Comment

From Microsoft, the Conditional Split transformation can route data rows to different outputs depending on the content of the data. The implementation of the Conditional Split transformation is similar to a CASE decision structure in a programming language. The transformation evaluates expressions, and based on the results, directs the data row to the specified output. This transformation also provides a default output, so that if a row matches no expression it is directed to the default output.

You can configure the Conditional Split transformation in the following ways:

Provide an expression that evaluates to a Boolean for each condition you want the transformation to test.
Specify the order in which the conditions are evaluated. Order is significant, because a row is sent to the output corresponding to the first condition that evaluates to true.
Specify the default output for the transformation. The transformation requires that a default output be specified.

Let’s take a look at how this transformation might be used in the real world.

Open Visual Studio and drag a Data Flow task into the design pane. Open the Data Flow task and drag in an OLE DB Source task. For this post, I’m going to use the AdventureWorks2019 database and the HumanResources.vEmployeeDepartment view.

This view has some good data to play around with, but we’re going to focus on the Department and Start Date columns. Let’s pretend the bossman needs to see all of the Employees in the Quality Assurance (QA), Production and Sales Department in a separate database table. Bonus, he needs to see all of the Production employees split up into two more tables based on who started before and after Jan 1, 2010. All other employees can go into their own table. Got it? Great! That’s 5 total tables. QA=1, Production=2, Sales=1, Leftovers=1 Let’s go.

Back in Visual Studio, drag in a Conditional Split task and connect it to our OLE DB Source.

Open the Conditional Split task editor and you’ll see a few options (from left to right, top to bottom):

We can use columns and/or variables and parameters in our expressions that define how to split the data flow.
We can use functions such as Date/Time, NULL and String in our expressions that define how to split the data flow.
These are the conditions that define how to split the data flow. These need to be set in priority order; any rows that evaluate to true for one condition will not be available to the condition that follows.

Let’s start adding some conditions for our data. First, we’ll add a condition for all of our QA Department Employees. I’ll name the output “QA” and my condition is pretty simple whereas Department == “Quality Assurance”.

I’ll do the same for Production, Sales and Leftovers (everything else that doesn’t satisfy a condition). Since Leftovers is everything else we’ll just change the name of the Default Output name to identify it.

Let’s go ahead and add our destination tasks (except for Production since we need another condition) and link them to the appropriate condition. See below for QA as an example. When we drag our connector to our destination task we get prompted with an Input Output Selection box. Here is where we choose our Condition that will match up with our table. For the screenshot below, we’ll choose QA output for our QA destination.

Now that we have QA mapped, go ahead and map Sales and Leftovers.

Looks great! QA, Sales and Leftovers are mapped successfully. Let’s take a look at adding another Conditional Split task for Production. Drag a Conditional Split task into the design pane and connect it to the current Conditional Split. It automatically maps to Production since it’s the only output left.

From our new Conditional split task, let’s open the editor and configure the date conditions for Production. We’ll leave the Default output name box as is since we shouldn’t have any leftover data from this split.

Now we can map the two new conditions to their appropriate destinations.

Cross fingers and hit Execute.

Yay, no red X! Let’s take a look at our SQL tables to make sure everything exported correctly.

Boom. Let’s go grab a bourbon!

SSIS Series: How to use SSIS Balanced Data Distributor

By bradyupton in Business Intelligence, Integration Services (SSIS), SSIS Series January 14, 2023 0 Comment

From Microsoft, the Balanced Data Distributor (BDD) transformation takes advantage of concurrent processing capability of modern CPUs. It distributes buffers of incoming rows uniformly across outputs on separate threads. By using separate threads for each output path, the BDD component improves the performance of an SSIS package on multi-core or multi-processor machines.

The Balanced Data Distributor transformation helps improve performance of a package in a scenario that satisfies the following conditions:

There is large amount of data coming into the BDD transformation. If the data size is small and only one buffer can hold the data, there is no point in using the BDD transformation. If the data size is large and several buffers are required to hold the data, BDD can efficiently process buffers of data in parallel by using separate threads.
The data can be read faster than the rest of the data flow can process it. In this scenario, the transformations that are performed on the data run slowly compared to the rate at which data is coming. If the bottleneck is at the destination, the destination must be parallelizable though.
The data does not need to be ordered. For example, if the data needs to stay sorted, you should not split the data using the BDD transformation.

Let’s dive in and take a quick look at how the Balanced Data Distributor works.

Go ahead and open Visual Studio, it might take a minute to load.

We’re going to use AdventureWorks2019 database for this example. The SalesOrderDetail table has over 121k records so that’s a good candidate.

SELECT * FROM Sales.SalesOrderDetail

Now that we have data let’s go back over to Visual Studio and see if it’s still spinning.

Drag in a Data Flow task, double click to open and then let’s drag in an OLE DB Source and a Flat File Destination task.

I’m going to configure the source to point to my AdventureWorks2019 database and the destination to point to a .csv file on my local laptop.

Easy enough. Let’s go ahead and run this and see what happens.

We can see that our 121k rows were written to our CSV file and the CSV file ended up being 12.3 MB. That’s pretty large for an email file attachment and might crash your laptop even trying to open this file. BDD not only offers performance benefits by using multiple threads, but it can also break up large files into smaller ones. IMO, this is what makes this task great.

We need to get this file down to less than 4MB so we’ll need to break this up 4 times. With that said, let’s add the BDD task between the source and destination tasks. There is nothing configurable *in* this task, however, there are some properties that may need to be tweaked. After adding the BDD task, we’ll need to add 3 more flat file destination tasks and 3 more flat file connection managers. End result should look something like this:

That’s really about it. Let’s fire it off and see the results of our flat files.

It worked! 121k rows were written across 4 flat files. Our flat file size is less than 4MB each and we can send these very easily via email. This is a very quick and easy way to split data between files, however, there is no order to these records so an ORDER BY is not helpful here. If the goal is to separate data by a category or condition then you’ll need to use a Conditional Split task which I wrote about in this post.

SSIS Series: How to use SSIS Aggregate Data Flow Task – Basic and Advanced

By bradyupton in Business Intelligence, Integration Services (SSIS), SSIS Series January 7, 2023 0 Comment

From Microsoft, the Aggregate transformation applies aggregate functions, such as Average, to column values and copies the results to the transformation output. Besides aggregate functions, the transformation provides the GROUP BY clause, which you can use to specify groups to aggregate across.

The Aggregate transformation supports the following operations.

Operation	Description
Group by	Divides datasets into groups. Columns of any data type can be used for grouping. For more information, see GROUP BY (Transact-SQL).
Sum	Sums the values in a column. Only columns with numeric data types can be summed. For more information, see SUM (Transact-SQL).
Average	Returns the average of the column values in a column. Only columns with numeric data types can be averaged. For more information, see AVG (Transact-SQL).
Count	Returns the number of items in a group. For more information, see COUNT (Transact-SQL).
Count distinct	Returns the number of unique nonnull values in a group.
Minimum	Returns the minimum value in a group. For more information, see MIN (Transact-SQL). In contrast to the Transact-SQL MIN function, this operation can be used only with numeric, date, and time data types.
Maximum	Returns the maximum value in a group. For more information, see MAX (Transact-SQL). In contrast to the Transact-SQL MAX function, this operation can be used only with numeric, date, and time data types.

This transformation could also be implemented with a SQL query after an initial load, but this task makes it nice to take care of the transformation within SSIS so that the data is ready on export. Let’s take a quick look at how it works.

Using AdventureWorks database, I’ll run the following query which gives me everyone in the Person.Person table that is located in the US.

SELECT firstname + ' ' + lastname AS PersonName, AddressLine1, AddressLine2, City ,sp.[Name] AS STATE, PostalCode, CountryRegionCode FROM person.Person p JOIN person.BusinessEntityAddress bea ON p.BusinessEntityID = bea.BusinessEntityID JOIN person.Address a ON bea.AddressID = a.AddressID JOIN person.StateProvince sp ON a.StateProvinceID = sp.StateProvinceID WHERE CountryRegionCode = 'US'

For this post, let’s pretend our boss wants to see how many people we have in each state. So a simple COUNT(*) and GROUP BY State query would work, but we’ll use SSIS to export this into a table using the Aggregate task.

Fire up Visual Studio and let’s create a Data Flow Task, add source connection and destination connection.

Next, let’s add our Aggregate task between our Source and Destination and open the editor:

This example is pretty straightforward as we want to do a COUNT on all records so we’ll choose (*) under Input Column, give it a name under Output Alias and the only available option for * is Count All.

Next, we want to get the number of people per state, so we’ll add State to our Input Column, give it a name, and select GROUP BY for our operation:

We can use multiple options on this screen such as SUM, AVG, MIN, MAX, etc. depending on data types.

Hit OK on the screen which will take us back to our Data Flow task. Taking a look at our Destination task, you can see I have a table named EmployeesPerState with only two columns, Employee and State:

Save the package and execute and you can see that we’ve inserted the data we need. Boss is happy and gives you a $6 bonus.

For Advanced Mode, you can create more than 1 GROUP BY criteria. For example, the boss now wants to see the number of people in each state, but he also wants to see how many people per zip code. For this, we would expand the Advanced/Basic drill down and add another input for his request. We’ll name the Aggregation PostalCode and Group by PostalCode. Like we did for our State aggregation, we’ll add a Count all Operation and (*) Input.

Back on the Data Flow task, we’ll drag another Destination task into the editor and configure.

You can now see both Destination tasks with a label for each.

If we run the package again, you can see results have been inserted into our PostalCode table and the boss gives you another $6 bonus and let’s you take Sunday off.

SSIS Series: Insert/Update a SQL Server table using Merge Join in SSIS

By bradyupton in Business Intelligence, Integration Services (SSIS), SSIS Series April 25, 2020 0 Comment

I was asked a question awhile back on the easiest way to do an incremental load from one SQL Server table into another SQL Server table. This is common in data warehouses or reporting tables as you wouldn’t want to truncate a large table and perform a full insert. Instead, you would want to only copy the changes from the source to the destination. This could be an insert or an update. The easiest way, IMO, is SSIS. Let’s take a look.

Here’s how my environment is setup. I have two databases appropriately named Transactional (for transactional data) and Reporting (for static report data). Two tables, Source and Destination.

I also added some dummy data as you see below. I’ve highlighted what is different and will need to be changed in the destination table.

Let’s open SSIS and create a new SSIS project.
Drag and drop a Data Flow task into the design window, right click, Edit:

Next, drag and drop two OLE DB source components into the design window. I’m going to rename mine Source and Destination to match my table names.

Configure the two OLE DB sources to match the source table and the destination table. Below is a screenshot of my source connection manager.

Once your connection managers are configured let’s drag and drop two Sort components below each OLE DB Source and connect them to each source. We will be using a merge join next and the merge join component needs to have data sorted in ASC or DESC order. I always choose to sort on the Primary Key for each table.

In my example, I’ll choose to sort on my primary key, ID.

Next, drag the Merge Join transformation into the Design window and drag the data path from the Sort component to the Merge Join. When you attach the arrow to the transformation, the Input Output Selection dialog box appears, displaying two options: the Output drop-down list and the Input drop-down list. The Output drop-down list defaults to Source Output, which is what we want. From the Input drop-down list, select Merge Join Left Input, as shown below. We’ll use the other option, Merge Join Right Input, for the other connection.

Next, connect the data path from the other Sort component to the Merge Join transformation. This time, the Input Output Selection dialog box does not appear. Instead, the Input drop-down list defaults to the only remaining option: Merge Join Right Input.

Now, let’s configure the Merge Join transformation.

The first setting in the Merge Join Transformation Editor is the Join type drop-down list. From this list, you can select one of the following three join types:

Left outer join: Includes all rows from the left table, but only matching rows from the right table. You can use the Swap Inputs option to switch data source, effectively creating a right outer join.
Full outer join: Includes all rows from both tables.
Inner join: Includes rows only when the data matches between the two tables.

For our example, we want to include all rows from left table but only rows from the right table if there’s a match, so we’ll use the Left outer join option.

You now need to select which columns you want to include in the data set that will be outputted by the Merge Join transformation. For this exercise, we’ll include all columns. To include a column in the final result set, simply select the check box next to the column name in either data source.

Almost finished, but first let’s add a Conditional Split transformation. This will allow us to insert new records or update previous records.

In the Conditional Split Editor, I created two outputs. If (Destination) ID is NULL then the record doesn’t exist so we’ll perform an INSERT. If the ModifiedDate is different between the two tables then we know something has been updated since the last execution and we need to update the record. See below.

Since we can perform and INSERT or an UPDATE we’ll need two destinations. First, for the INSERT, we’ll simply be doing an INSERT into the table so we can drag the OLE DB Destination component into the window, choose “INSERT” in the Input Output selection window, and use the Reporting connection manager.

I’m going to check the “Keep Identity” box since the ID column is an identity column.

Next, for the UPDATE statement we’ll drag the OLE Command component into the window and configure it. Select “UPDATE” in the Input Output Selection window.