The Nutanix Solution

The Nutanix platform includes up to four individual server nodes housed in a space-efficient, 2U rack-mountable appliance. Each server node is equipped with best-in-class Intel processors and high-speed memory for the most demanding workloads.

You can add additional Nutanix appliances to an existing cluster to scale performance and capacity. Or, small clusters can be combined with larger clusters to streamline administration.

Virtualization hypervisors are pre-installed, ready for the rapid deployment of virtual machines (VMs). Each node can support just one or up to 100 virtual desktops.

To optimize performance further, the Nutanix solution manages storage resources in tiers. The data requested most frequently, the “hot” data, moves to the highest-performance tier, which uses the PCIe SSD. Unlike competing solutions, this PCIe-based flash tier serves as a true persistent tier, not simply a cache.

Heat-optimized tiering (HOT) ensures the highest performance for both write and read operations. When a piece of data is accessed less frequently, the Nutanix software moves the data to a lower-cost, higher-capacity storage medium such as SATA-HDDs.

The proliferation of VMs throughout enterprise datacenters is driving I/O requirements higher and higher. The need for greater performance in the storage tier has meant a costly expansion of legacy arrays (e.g. more spindles). The result is extreme underutilization of expensive storage hardware.

Nutanix solves this problem by allowing storage performance to increase incrementally, one node at a time.

The Nutanix software architecture pools storage resources across physical nodes, and manages them as a single namespace. When you need more performance or capacity, you can add Nutanix nodes one at a time and precisely match storage capacity and I/O performance with VM deployments.

Every VM seamlessly accesses physical-storage resources across all nodes. Newly provisioned VMs automatically leverage the pooled storage as their storage target, so they are ready for immediate operation. You do not have to manually define storage volumes, LUNs, or RAID groups when you provision a new VM. The Nutanix architecture delivers advanced, enterprise-class storage capabilities, such as snapshots, cloning, DR, compression, and error detection. The next-gen converged design eliminates the need for a separate SAN infrastructure.

Enterprises no longer need to make expensive, step-function increases in storage-array investments just to keep pace with rapid virtualization. The inherent flexibility of Nutanix’s software-defined architecture ensures that customers continue to get value from functional hardware. The heat-optimized data tiering gives customers
the best possible price-performance ratios for their investment.

Traditional architectures use complex networks (SANs or NAS) to transfer data between centralized storage arrays and server resources. The constant movement of data over a network leads to significant inefficiencies and performance degradation.

The complexity of maintaining dedicated storage networks further increases datacenter costs, and inhibits agility and elasticity of the infrastructure. LUN based architectures require protection against disk resource contention and data corruption with functionality such as zoning and LUN masking.

Converging compute (server) and storage onto a single platform keeps data local to the virtual machines, where it is needed. This system eliminates expensive and complex SANs, with no loss of critical data-management and data-optimization capabilities.

Virtualization teams and application owners can rapidly provision new virtual workloads, without having to worry about the constraints of a separate storage tier. The result is faster time-to-value for new workloads and far lower operating costs.

The Nutanix distributed file system (NDFS) incorporates the same principles and uses many of the technologies the massive Google File System (GFS) uses as well as big-data techniques such as MapReduce.

Nutanix adapted these technologies for mainstream enterprises by adding support for virtualization.

The Nutanix software architecture provides nearly unbounded linear scale-out. As your company grows, you can add compute and storage one node at a time. The Nutanix architecture enables single clusters to be built with nodes configured to match specific workloads: Some might be compute-intensive and others might need more storage capacity.

This flexible approach contrasts to first-generation converged systems that deliver vendor-bundled systems with legacy storage systems in a fully populated rack. The result is the same bottlenecks and performance limitations of SAN or NAS storage.

Dynamic cluster technology enables self-discovery of new Nutanix nodes, which can be added without disrupting existing cluster resources. The technology automatically adds new compute and storage capacities, and rebalances resource allocation across the cluster for maximum efficiency.

The Nutanix architecture is hypervisor-agnostic, and works with multiple virtualization technologies. All VM-level capabilities are delivered independently, including high availability (HA), disaster recovery (DR), and more.

Virtualization teams are now flexible enough to host any VM using their hypervisor of choice. They are not locked into one virtualization vendor. A single Nutanix cluster can support different hypervisors simultaneously, with a nearly unlimited ability to support mixed workload environments. This allows the Nutanix-enabled datacenter to serve multiple IT needs across different business units at the same time.

Virtualization teams can continue to use existing management and reporting tools so there is no disruption to ongoing operations. The management of all shared storage is completely abstracted from the virtualization stack: Each guest VM operates as if it is working directly with a traditional datastore.