HEVC media optimization

Do you have a large collection of video media files not using HEVC (H265) yet? There is a massive amount of disk space coming your way if you flick over to the new video codec format.

HEVC definitely lives up to its name, for most media you can expect a 70% or more disk savings from transcoding from an old codec. There are some catches though… If you want your TV to play it direct (i.e. straight off the file) the codec will need to be supported by it. You can of course get around this by using a media server such as Plex or Emby which will transcode from HEVC back to a compatible format.

Why would you transcode to HEVC? – again, disk space. HEVC as stated above can reduced you Media footprint significantly. You could boost your quality and save your disk space at the same time by recording at a higher resolution then applying the HEVC codec.

I created a powershell script to transcode my media to HEVC using my AMD graphics card. The advantage of doing this is that transcoding completed by my GPU is significantly faster than my CPU. I do not have the graphics card in my media server, so instead connect via SMB and let my gaming machine run the transcoding from remote…

The powershell script uses ffmpeg to ;

  • transcodes video stream to hevc using AMD h/w encoder
  • copys all existing audio and subtitles (i.e. no conversion)
  • works in batches (to prevent constant scanning of files) – able to set max batch size and processing time before re-scanning disk
  • overwrites source with new HEVC transcode if move_file = 1 (WARNING this is default!)
  • checks to see if video codec is already HEVC (if so, skips)
  • writes transcode.log for successful transcode (duration and space savings)
  • writes skip.log for already hevc and failed transcodes (used to skip in next loop, errors in transcode.log)

Check here for updates and script – https://github.com/dwtaylornz/hevctranscode

windows 2012 – dedup

If you have some archive data that you would like to copy onto a  win 2012 server, but dont know how to estimate the end volume size (if VM etc). Start small – about 1/2 of the total capacity of the drive. Robocopy the data onto the drive until nearly full, manually run dedupe, wait until complete, continue robocopy, repeat and rinse. There will be occasions where you will run out of disk space at that point its time to extend your disk.

You will need to enable deduplication on the file server, either hit the GUI or use the following to enable the module;

Import-Module ServerManager
Add-WindowsFeature -name FS-Data-Deduplication

The powershell commands that you’ll use;

Enable-DedupVolume E:
(enables E drive for deduplication)

Set-DedupVolume E: -MinimumFileAgeDays 0
(default is 5 days – which may be sufficient for achival type files)

Start-DedupJob E: –Type Optimization
(manually starts dedup on E drive)

Get-DedupJob
(shows status of currently running dedup and scheduled dedup jobs)

I have used the following powershell script to continually run the dedup process on my archive drive (during migration work). Checks if dedup processes are running – if not execute dedup on archive drive.


cls
$x = 1
while ($x = 1)
{
$dedupcheck = Get-Dedupjob | Out-String;
if ($dedupcheck -eq "") {
echo "starting dedup process on F drive";
Start-DedupJob F: -type Optimization;
};
echo "dedup jobs running - sleeping for 2 min";
sleep 120;
echo "";
get-dedupstatus;
echo "";
}

powershell – move machine types to certain OUs in active directory


-------------- select additional parameters ----------------------

Get-ADComputer -Filter * -Properties ipv4Address, OperatingSystem, OperatingSystemServicePack | Format-table name, ipv4*, operatingsystem

-------------- move script -----------------

[cmdletbinding()]

param (
[parameter(mandatory=$true)]
$TargetOU
)

Import-Module ActiveDirectory
$Domain = [ADSI]""
$DN=$domain.distinguishedName
$SourcePath = "CN=Computers," + $DN
$Computers = Get-ADComputer -Filter * -SearchBase $SourcePath
if(!$Computers) {
write-host "No Computers are found in default container"
return
}
foreach ($Computer in $Computers) {
if(!(Move-ADObject $Computer -TargetPath $TargetOU)) {
$Status = "SUCCESS"
} else {
$Status = "FAILED"
}
$OutputObj = New-Object -TypeName PSobject
$OutputObj | Add-Member -MemberType NoteProperty -Name ComputerName -Value $Computer.Name.tostring()
$OutputObj | Add-Member -MemberType NoteProperty -Name SourcePath -Value $SourcePath
$OutputObj | Add-Member -MemberType NoteProperty -Name DestinationPath -Value $TargetOU
$OutputObj | Add-Member -MemberType NoteProperty -Name Status -Value $Status
$OutputObj
}