Code Library

Storage Spaces Direct enables service providers and enterprises to use industry standard servers with
internal drives to build highly available and scalable software defined storage. Using servers with internal drives decreases complexity, increases scalability, and enables use of storage devices that
were not previously possible, such as SATA solid state disks to lower cost of flash storage, or NVMe
solid state disks for better performance.

Storage Spaces Direct removes the need for a shared SAS fabric, simplifying deployment and configuration. Instead, it uses the network as a storage fabric, using SMB3 and SMB Direct (RDMA) for high-speed, low-latency CPU-efficient storage. To scale out, simply add more servers to increase
storage capacity and I/O performance. Following are some more features and characteristics of Storage Spaces Direct:

 Storage for Hyper-V and Microsoft Azure Stack The primary use cases for Storage Spaces Direct is as storage for Hyper-V VMs or as storage for Azure Stack.

 Hardware Storage Spaces Direct makes it possible to build highly available and scalable storage solutions using modern storage hardware like SATA SSD for lower cost and NVMe SSD for better performance and less CPU overhead. It can also use RDMA-enabled network infrastructure for low-latency storage with less CPU overhead than traditional Ethernet. Less CPU overhead means increased workload density.

 Prescriptive configurations Microsoft is working closely with its hardware partners to define and validate prescriptive server configurations for Storage Spaces Direct. Using these server configurations provides the best possible experience with Storage Spaces Direct with the full feature set and best performance.

 Storage Configurations You can use Storage Spaces Direct with various storage configurations. The most common configurations are:
    SSDs with traditional hard drives, where the SSDs are used as a read/write cache to accelerate I/O performance.
    All-flash configuration with NVMe SSDs and SATA SSDs for extremely high I/O performance.
    Three tiers of physical storage, NVMe SSDs, SATA SSDs, and traditional hard drives.

 Deployment Choice Storage Spaces Direct provides customers with a deployment choice, either as hyper-converged infrastructure or as converged infrastructure. In a hyperconverged infrastructure, compute and storage resources are provided by the same machines, simplifying scale and management. In a converged infrastructure, compute resources are separate from storage resources, allowing for increased scalability and independent scaling of compute and storage.

 Fault Tolerance Storage Spaces Direct is resilient to drive failures. When drives fail, degraded data is automatically reconstructed on the remaining drives. Storage Spaces Direct supports three fault domain types: a) server, b) chassis, and c) rack, and all data placement, data repair, and data rebalancing will adhere to the fault domain configuration.

 Accelerated Erasure Coding Storage Spaces Direct introduces hybrid volumes, which is in addition to the existing mirror and erasure coding volume types. Hybrid volumes mixes the best of mirror (performance) with the best of erasure coding (efficiency) into a single volume with automatic real-time storage tiering.

 Efficient VM check-points Storage Spaces Direct utilizes the new ReFSv2 file system, which when combined with Hyper-V can do very quick and efficient VM checkpoints with little overhead and storage I/O.

 Scalability Storage Spaces Direct can scale from 2 to 16 servers. You can add servers as needed, and data can be rebalanced to best utilize the additional resources. Microsoft and Intel demonstrated a 16 server Storage Spaces Direct deployment using all NVMe SSDs at IDF 2015.

 Health Service Storage Spaces Direct includes an intelligent built-in diagnostic engine that makes it possible for administrators with limited technical expertise to monitor and operate the system day to day.
    The Health Service actively monitors the underlying cluster, storage hardware, and software defined storage stack to detect problems and generate alerts that contain precise instructions for how to react.
    Performance and capacity information is aggregated to present a holistic, high-level view of available resources.
    Frequent tasks such as drive replacement and drive firmware updates are automated to reduce the burden on the administrator.

Source of Information : Microsoft Introduction Windows Server 2016