Data deduplication

Windows Server 2016 with Hyper-V offers built-in deduplication at no extra charge. It's a great way to reduce your storage capacity footprint with very little configuration. However, data deduplication still comes at a price-it requires additional I/O capacity. Therefore, on a general use file server, it will not affect hot data until it's reached a certain file change age. Besides the I/O-hit, volumes with active deduplication will fragment more easily causing single file operations to take longer on deduped volumes. Hyper-V takes some precautions to avoid a big performance hit, that is, every block referenced more than 100 times will be written a second time.

Real-life experiences tell us that the overall gain in saved space outweighs the performance cost on file servers, library servers, and VDI Hyper-V hosts. Running Windows data deduplication on running VMs with server workloads is not supported. Before using deduplication, you can test how much space saving dedup...