MapReduce Tiering: Intelligent Distributed Data Tiering
Flash storage is used both as a cache and as a persistent data tier in a Nutanix system. Data is intelligently placed in the optimal storage tier — flash or HDD — to yield the fastest possible performance. MapReduce tiering technology ensures that the most frequently accessed data is available in the SSD or cache tier. As data becomes ‘cold,’ it is demoted into the higher capacity HDD tier. This is ensures that SSD and cache capacity remains available for new ‘hot’ data.
Nutanix incorporates the powerful Snappy compression algorithm to increase the effective storage capacity of the system up to 4X. Unlike traditional storage solutions that perform compression for entire LUNs or disks, Nutanix compresses data at the sub-block level for increased efficiency and greater simplicity. Chunk sizes up to 128 KB enable very high compression ratios.
To learn more about the benefits of Compression, download the solution brief.
MapReduce Compression: Distributed Post-process Compression
Administrators can set policies to run the Nutanix compression feature post-process to eliminate any performance impact on the write path. MapReduce Compression is distributed across all nodes in the cluster and scales out as the cluster grows.
Elastic Deduplication: Fine-grained Inline and Post-process Deduplication
Nutanix systems offer two flavors of data deduplication, together called Elastic Deduplication – Inline Performance Deduplication in the content cache (SSD and memory) reduces the cache footprint of the application working set yielding significant performance improvements, while global, post-process MapReduce Deduplication in the capacity tier increases the effective storage capacity of a cluster.
Data is fingerprinted using a SHA-1 hash on ingest, with fingerprint information stored in the Nutanix metadata layer. Deduplication operations are software-driven, and leverage the hardware-assist capabilities of the Intel chipset for SHA-1 hash generation for the fastest possible performance.
For applications with large common working sets, such as virtual desktop infrastructure (VDI) deployments, inline deduplication increases effective flash and memory resources by up to 10x and delivers nearly instantaneous application response times. MapReduce Deduplication is global and distributed across all nodes in the cluster, effectively minimizing any performance overhead.
For a more in-depth overview of Elastic Deduplication, download the technical guide.
Read the MapReduce Deduplication Capacity Summary to understand how much capacity optimization to expect
Heterogeneous Clusters: Mix and Match Nutanix Appliances
Customers can mix different models of Nutanix appliances with different ratios of storage and compute in a single cluster. This allows them to effectively scale compute and storage independently in a cluster depending on growing needs.
Thin Provisioning for Storage Efficiency
Storage for virtual machines is thinly provisioned in the system. Administrators can set the capacity of a vDisk, but physical storage is allocated only when required. Administrators can also set a minimum reservation parameter that guarantees the specified amount of storage for a collection of vDisks. This provides for less overprovisioning of storage and more granular control to the administrator.
Flexible Block Size for Performance and Metadata Efficiency
End-to-end support for a wide range of I/O sizes, from 512 bytes up to 1 MB ensures optimal performance across wide range of workloads and delivers application-tuned metadata efficiency. Unlike traditional storage systems that may need to break up and store large incoming I/O requests into 4KB chunks, Nutanix can handle block sizes of variable length across the entire write path.
Shadow Clones: Distributed Localized Caching of VM Data for Performance
Shadow Clones enable distributed caching of VM data in a multi-reader scenario. VDI deployments, where many linked clones forward read requests to a central master (base VM), is a prime example. In the case of VMware View, this is called the replica disk and is read by all linked clones. Similarly, in Citrix XenDesktop deployments this is called the MCS Master VM. Shadow Clones improve performance in nearly any multi-reader scenario (e.g., deployment servers, repositories, etc.).
With Shadow Clones, Nutanix actively monitors vDisk access trends. If there are requests originating from more than two remote CVMs, as well as the local CVM, and all of the requests are read I/O, the vDisk will be marked as immutable. Once the disk has been marked immutable, the vDisk is then cached locally by each CVM so read operations are now satisfied locally by direct-attached storage resources.
To learn more about Shadow Clones, download the tech guide.