This post was authored by the Microsoft Mechanics team.
On the latest Microsoft Mechanics we take a look at software defined storage using new capabilities in Windows Server 2016. Elden Christensen from the Windows Server storage team joins Matt McSpirit to expose what’s behind the black box of a typical Storage Area Network (SAN) and how you can easily and inexpensively build out your own highly available and scalable storage solutions using Windows Server 2016 and Storage Spaces Direct with commodity hardware and hard drives.
Storage Spaces Direct integrates with the Windows Server software defined storage stack, including Scale-Out File Server, Clustered Shared Volume File System (CSVFS), Storage Spaces and Failover Clustering.
Deployment Configuration Options for Storage Spaces Direct
There are two primary configuration options for Storage Spaces Direct and Elden describes both on the show: converged and hyper-converged. Both options provide storage for Hyper-V workloads.
The converged deployment option has the Hyper-V servers in a separate cluster from the Storage Spaces Direct servers. Virtual machines are configured to store their files on the Scale-Out File Server which is accessed through the network using SMB3 for local-like performance. This allows you to scale Hyper-V clusters (compute) and Scale Out File Server cluster (storage) independently.
The hyper-converged deployment option has the Hyper-V workload (compute) and Storage Spaces Direct (storage) components on the same cluster. Virtual machine files are stored and run on the local clustered shared volume without the need to configure a Scale-out File Sever.
When deciding which option works best for your situation, remember hyper-converged has less complexity because of the unified compute and storage, while converged provides more flexibility to scale storage and compute independent of one another. Large scale infrastructures will typically use converged to allow for the independent scaling, whereas smaller systems can benefit from the simplicity of hyper-converged.
Elden and Matt explain deployment configuration options and even go under the hood of Storage Space Direct to describe how it works to optimize performance and resiliency using caching, mirroring and erasure coding.
Storage Replica
Storage Replica is a new capability in Windows Server 2016 that you can use to protect your data against catastrophic losses. Storage Replica synchronously copies your data across distances from rack-scale to geographic scale. With Storage Replica you can switch workloads among locations as a proactive action or in response to catastrophes — without risking data loss. You can stretch or replicate clusters, so that workloads can be run in multiple datacenters for quicker data access by local users and applications, as well as better distribute load and usage of compute resources.
Try it!
You can try out Storage Replica or Storage Spaces Direct in either converged or hyper-converged configurations today using the Windows Server 2016 Technical Preview and share your testing benchmarks with the community. Go to aka.ms/GetWindowsServer2016 and keep watching Microsoft Mechanics for the latest tech updates.