File size reported reported differently

Question

Can someone help me explain why this file size is reported differently depending whether you list file or folder properties:
This is an Azure Basic_A1 VM. Drive E: is standard VHD attached to the VM (Page Block Storage).
This is an active/open file that is currently being written to, and is growing in size.
How is it that file size is reported as 8 GB if you query the folder, and 24 GB if you query the file?
Sam Boutros, Senior Consultant, Software Logic, KOP, PA http://superwidgets.wordpress.com (Please take a moment to Vote as Helpful and/or Mark as Answer, where applicable) _________________________________________________________________________________
Powershell: Learn it before it's an emergency http://technet.microsoft.com/en-us/scriptcenter/powershell.aspx http://technet.microsoft.com/en-us/scriptcenter/dd793612.aspx

Answer 1

I ran this script for the last 8 or so hours out of this 19 hour upload:
$Folder = "e:\Backups\Sam Client\To Azure Page Block Repository"
$Log = "c:\Sandbox\Upload4.csv"
Write-Output '"FileQuery","FolderQuery","Time"' | Out-File $Log
Do {
$a = (dir "$Folder\To Azure Page Block Repository2014-09-23T051450.vbk").length/1GB
$b = (dir $Folder)[1].length/1GB
cls
Write-Host "File size reported from file list: $('{0:N2}' -f $a) GB" -ForegroundColor Green
Write-Host "File size reported from folder list: $('{0:N2}' -f $b) GB" -ForegroundColor Yellow
dir $Folder | FT -AutoSize
Write-Output """$a"",""$b"",""$(Get-Date -Format 'hh:mm:ss tt')""" | Out-File $Log -Append
Start-Sleep -Seconds 10
} while ($a -ne $b)
That collected over 3,000 sampling points that are 10 seconds apart. I then graphed that data:
The vertical axis is file size in GB, the horizontal axis is time (not directly - it's sampling points)
And you can see file size as reported by folder query lags behind file size as reported by file query. The amount of lag was upwards of 90GB at its highest point. And it took anywhere from roughly 5 minutes to 3 hours for 1 indicator to catch up with the
other. 
The question is why these two sets of numbers are different? Is there caching going on somewhere? Is one indicator accounting for all file blocks while the other is accounting for committed blocks only? Can someone from Microsoft shed some light on this?
This is important to know to be able to plan disk space requirements when attempting to use Azure as backup target with large data sets regardless of the software used.
Sam Boutros, Senior Consultant, Software Logic, KOP, PA http://superwidgets.wordpress.com (Please take a moment to Vote as Helpful and/or Mark as Answer, where applicable) _________________________________________________________________________________
Powershell: Learn it before it's an emergency http://technet.microsoft.com/en-us/scriptcenter/powershell.aspx http://technet.microsoft.com/en-us/scriptcenter/dd793612.aspx