Update Standard.json for OsDisk

[jasbryan]

Modified OS disk to SSD drive.

[jasbryan]

It is working now. Please give it a try and let me know what you think.

[ecalder6]

Gave it a shot. I'm not sure if the VM is using an SSD based on the benchmark below. What do you think?

[jasbryan]

Looks pretty badass to me. I have been running it with the SSD for a few weeks. Runs great. You really need it for big games. I play Star Citizen and it is going to the drive all the time. I wish I could force more into RAM since these VMs have so much.
There are even faster SSD options available but I have not compared them.

[jasbryan]

Meant to request merge not to close it.. I am new to GitHub..

[ecalder6]

No worries. I'll test a bit more and merge it.

[ecalder6]

Here's the speed I got with the current code:

I don't think changing the storage account type to StandardSSD_LRS changes the disk to SSD. I'm pretty sure you'll have to deploy NVv2 VMs for SSD: https://azure.microsoft.com/en-us/updates/public-preview-new-gpu-enabled-nvv2-azure-vms-for-graphics-intensive-applications/

[ecalder6]

SSD is not deployed.

[jasbryan]

When I am running with this configuration my Disk shows standard SSD. IT is not the new, faster advanced SSDs but the standard ones.

[ecalder6]

My tests show that there isn't a difference in speed to justify the change. Feel free to do some speed comparisons yourself to show the performance gain.

[jasbryan]

I ran comparison tests with an HDD and an SSD mounted and there were noticeable speed increases.
Here is the comparisons.

MultipleDiskReport.zip

[ecalder6]

Thanks for the comparison! How did you setup your test environment? Are both C and E drive managed disks?

[jasbryan]

I beleive they are both managed disks. I just added the HDD drive and then ran the test on both. So the C drive was still running the OS during the test and was still much faster.

[ecalder6]

Nice. So just to confirm, C drive is the Standard SSD that's deployed using your change, and E drive is the Standard HDD that you've added through the portal/powershell?

If so, my test methodology is a bit different. I've deployed a VM using your changes and a VM using the current code and compared the speed of the two C drives using CrystalDiskMark. Could you try testing it this way (don't really care about the software used) to make sure the drive deployed with your change performs better than the drive deployed currently? I could also test again as well when I get a chance.