When using Microsoft Deployment Toolkit to Capture Reference Image for MDT or SCCM you know it can take quite some time to finish.
There is 3 ways to heavily reduce that time. In a worst case scenario you will be able to cut it in half, others are reporting 400% improvements! So let’s get started.
1. Use Multiple vCPUs
I guess out of habit we normally create VM’s with only 1 vCPU. Change that to 2 vCPUs on the VMs you use to Build and Capture Reference Image. That will almost cut the time in half.
2. Use DSIM instead of ImageX
When you use Microsoft Deployment Toolkit with the default scripts it uses ImageX for the image capture. ImageX has a bug that spends at least 15 minutes just to scan the file system before starting. You should switch to DSIM which are much faster. Click here to download the script.
3. Use RAM Disk
If the VMs you use to Capture Reference Image is hosted on SSDs this won’t apply to you. But if you use SAN or Local SATA disks you can reduce the build and capture time by using a RAM Disk. You can use this free software from StarWind. Create the RAM disk and format it through Disk Management, don’t use the StarWind tool for that.
4 thoughts on “Reduce The Time To Capture Reference Image”
I think you’ve made a slight mistake on point 2 – you said that DSIM is used by default, but DSIM has a fault, so switch to DSIM instead; I’m guessing (based on the title) you mean it uses ImageX by default and ImageX has an issue so use DSIM instead?
Thanks Danny, just fixed it.
How do we use the RAM disk? Could you just explain a little bit more why the RAM disk is helpful and how it’s used?
Check this : http://www.starwindsoftware.com/high-performance-ram-disk-emulator. It´s all about speed, if you don´t have a SSD, use a RAM disk instead, SATA is way to slow.