When you think filtering in Powershell, the first thing that often comes to mind is the venerable Where-Object cmdlet. It's used quite often, and can be a very easy way to remove elements from a pipeline based on a set of conditions you supply. Consider this basic example:

PS> 1..10 | Where-Object {$_ -eq 7}

In an array of 10 objects, this task takes virtually no time to complete. We'll illustrate how to measure this performance when we do the same for 1,000,000 numbers:

PS> $Results = Measure-Command { 1..1000000 | Where-Object {$_ -eq 842} }

PS> $Results.TotalSeconds

We see it took over 6 seconds to complete. If this command were inside a foreach loop ( or even a nested foreach loop) it can quickly add up in terms of overall run time.

Enter the .FindAll() method.

FindAll is a very common method if you're familiar with C#, and it's implemented in a wide variety of classes. In this case we're going to look at the [Systems.Collections.Generic.List<t>] class. (You can read more about that at Microsoft's documentation here.) When we invoke this method with Powershell we're, in essence, creating a sort of function inside the constructor with a parameter and a script block to be used as search criteria for our filter. Here is the same above example using FindAll instead of Where-Object:

PS> $Results = Measure-Command {
    # Create the list
    $List = [System.Collections.Generic.List[Object]]::new(1..1000000)
    # Query the list for the number 842
    $List.FindAll({ param($item); $item -eq 842 })

PS> $Results.TotalSeconds

That's a savings of almost 50% for the same query filter!


This method isn't one you'll use every day, but under the right circumstances, it can be a powerful way to query a collection for specific criteria and improve your script performance.