Storage Spaces Direct uses industry-standard servers with local-attached drives to prepare highly available, highly scalable software-defined storage with a fraction associated with the of traditional SAN or NAS arrays. Its converged or hyper-converged architecture radically simplifies procurement and deployment, while features like caching, storage tiers, and erasure coding, and the latest hardware innovation like RDMA networking and NVMe drives, deliver unrivaled efficiency as well as.

Storage Spaces Direct is included in Windows Server 2016 Datacenter and Windows Server Insider Preview Builds.

Warning

Microsoft boasts a critical product advisory for Storage Spaces Direct customers making use of Intel P3x00 category of NVMe devices (all capacities of this P3500, P3600, P3700). See Expertise article 4052341 to acquire more information.

Understand

Overview (you were here)
Understand the cache
Fault tolerance and storage efficiency

Plan

Hardware requirements
Choose drives
Plan volumes

Deploy

Hyper-converged solution
Create volumes

Manage

Add servers or drives
Taking a server offline for maintenance
Remove servers
Extend volumes
Update drive firmware

Key benefits

Simplicity. Range from industry-standard servers running Windows Server 2016 to the first Storage Spaces Direct cluster inside Fifteen minutes. For System Center users, deployment is simply one checkbox.

Unrivaled Performance. Whether all-flash or hybrid, Storage Spaces Direct easily exceeds 150,000 mixed 4k random IOPS per server with consistent, low latency as a result of its hypervisor-embedded architecture, its built-in read/write cache, and support for cutting-edge NVMe drives mounted appropriate the PCIe bus.

Fault Tolerance. Built-in resiliency handles drive, server, or component failures with continuous availability. Larger deployments may possibly be configured for chassis and rack fault tolerance. When hardware fails, just swap versus eachother; the software heals itself, without the need of complicated management steps.

Resource Efficiency. Erasure coding delivers roughly 2.4x greater storage efficiency, with unique innovations like Local Reconstruction Codes and ReFS real-time tiers to improve these gains to hard disk drives and mixed hot/cold workloads, all while minimizing CPU consumption giving resources returning to where they're needed most - the VMs.

Manageability. Use Storage QoS Controls to maintain overly busy VMs at bay with minimum and maximum per-VM IOPS limits. The medical Service provides continuous built-in monitoring and alerting, and new APIs help it become straightforward collect rich, cluster-wide performance and capacity metrics.

Scalability. Raise to 16 servers additionally 400 drives, for as long as 1 petabyte (1,000 terabytes) of storage per cluster. To scale out, simply add drives or combine servers; Storage Spaces Direct will automatically onboard new drives get started with their company. Storage efficiency as well as improve predictably at scale.

Deployment options

Storage Spaces Direct was designed for a couple of distinct deployment options:
Converged

Storage and compute in separate clusters. The converged deployment option, often known as 'disaggregated', layers a Scale-out File Server (SoFS) atop Storage Spaces Direct to render network-attached storage over SMB3 file shares. This allows for scaling compute/workload independently within the storage cluster, necessary for larger-scale deployments for example Hyper-V IaaS (Infrastructure like a Service) for carriers and enterprises.

Hyper-Converged

One cluster for compute and storage. The hyper-converged deployment option runs Hyper-V virtual machines or SQL Server databases over the servers supplying the storage, storing their files concerning the local volumes. This eliminates the requirement to configure file server access and permissions, and reduces hardware costs for small-to-medium business or remote office/branch office deployments. See Hyper-converged solution using Storage Spaces Direct.
What's the deal

Storage Spaces Direct is going to be evolution of Storage Spaces, first introduced in Windows Server 2012. It leverages any number of the features you fully understand today in Windows Server, just like Failover Clustering, the Cluster Shared Volume (CSV) file system, Server Message Block (SMB) 3, ultimately Storage Spaces. On top of that introduces new technology, such as the Software Storage Bus.

Here's a review of the Storage Spaces Direct stack:

Networking Hardware. Storage Spaces Direct uses SMB3, including SMB Direct and SMB Multichannel, over Ethernet to share between servers. We recommend 10+ GbE with remote-direct memory access (RDMA), either iWARP or RoCE.

Storage Hardware. From 2 to 16 servers with local-attached SATA, SAS, or NVMe drives. Each server will require at any rate 2 solid-state drives, and a minimum of 4 additional drives. The SATA and SAS devices will have to be behind a host-bus adapter (HBA) and SAS expander. We highly recommend the meticulously engineered and extensively validated platforms in our partners (just around the corner).

Failover Clustering. The built-in clustering feature of Windows Server must be used to get the servers.

Software Storage Bus. The Software Storage Bus is totally in Storage Spaces Direct. It spans the cluster and establishes a software-defined storage fabric whereby the many servers understand every single piece of each other's local drives. Imaginable becoming replacing costly and restrictive Fibre Channel or Shared SAS cabling.

Storage Bus Layer Cache. The Software Storage Bus dynamically binds the fastest drives present (e.g. SSD) to slower drives (e.g. HDDs) to allow server-side read/write caching that accelerates IO and boosts throughput.

Storage Pool. The gathering of drives that form the first step toward Storage Spaces is the storage pool. It's automatically created, and all eligible drives are automatically discovered and added onto it. We highly recommend you buy one pool per cluster, when using the default settings. Read our Deep Dive to qualify for the Storage Pool pertaining to.

Storage Spaces. Storage Spaces provides fault tolerance to virtual "disks" using mirroring, erasure coding, or both. You can think of getting to be distributed, software-defined RAID utilizing the drives for the pool. Kept in storage Spaces Direct, these virtual disks normally have resiliency to 2 simultaneous drive or server failures (e.g. 3-way mirroring, with each and every data copy in the different server) though chassis and rack fault tolerance can also be available.

Resilient File System (ReFS). ReFS may be the premier filesystem purpose-built for virtualization. It has dramatic accelerations for .vhdx file operations which can include creation, expansion, and checkpoint merging, and built-in checksums to detect and proper bit errors. It also introduces real-time tiers that rotate data between so-called "hot" and "cold" storage tiers in real-time determined by usage.

Cluster Shared Volumes. The CSV file system unifies all of the ReFS volumes correct single namespace accessible through any server, to make certain that to every server, every volume works like it's mounted locally.

Scale-Out File Server. This final layer should be applied in converged deployments only. Top-quality remote file access with the SMB3 access protocol to clients, for instance another cluster running Hyper-V, on top of the network, effectively turning Storage Spaces Direct into network-attached storage (NAS).