All Articles

Azure D-Series VM Benchmark

Following the news of the D-series VM in Azure i wanted to test it out to see what kind of performance it delivered. I picked D1 with the following specs: azuredpic1 First of, the news of having a SSD intrigued me, so i wanted to test IOPS performance. For this i choose SQLIO and the powershell script by Mikael Nystrom. As it says on the website the powershell script uses SQLIO to perform IO tests on the selected disk. I started with testing the C: drive to get some baseline performance. [wpgist id=“ef368e9aafdffb912713”] azuredpic2 Next i ran the same test against the D: drive which is the one backed by SSD’s. azuredpic3 As expected we see a huge performance benefit from the SSD’s, only drawback is that they are non-persistent, from the Azure blog

On these new sizes, the temporary drive (D: on Windows, /mnt or /mnt/resource on Linux) are local SSDs. This high-speed local disk is best used for workloads that replicate across multiple instances, like MongoDB, or can leverage this high I/O disk for a local and temporary cache, like SQL Server 2014′s Buffer Pool Extensions. Note, these drives are not guaranteed to be persistent. Thus, while physical hardware failure is rare, when it occurs, the data on this disk may be lost, unlike your OS disk and any attached durable disks that are persisted in Azure Storage.

From this i read that the only scenario where i will lose data is in the case of hardware failure, how offen that may happen is yet to be determined. Microsoft is being conservative and saying you should only use it for replicated workloads, but perhaps some crafty ITPRO’s will find ways to uttilize the power of the SSD’s without the chance of loosing data.

Published 28 Sep 2014