Issue link: https://insights.oneneck.com/i/1380812
18 • Auto-leveling. Advanced algorithms ensure that data is balanced across nodes. Data from any virtual machine can utilize storage on other nodes when needed, eliminating the possibility of running out of storage on any single node and simplifying data management. If a particular node has more hot data than it has local SSD, available SSD capacity on other nodes is used. • Auto-archiving. A Nutanix cluster can include capacity- only nodes to increase available storage. When these nodes are deployed, cold data preferentially finds its way to them. In effect, this provides auto-archiving of cold data. Whenever data becomes active again, it moves back to the node where it is needed without intervention. • Replication factor. Nutanix Enterprise Cloud Platform relies on a replication factor (RF) for data protection and availability. Either two or three copies of all data are maintained on different nodes within a cluster. Nutanix's patented erasure coding algorithm, EC-X, can be used to reduce the storage overhead resulting from RF. Using EC-X can increase usable space in the cluster by up to 70%. • Self-healing. Replication factor also enables a Nutanix cluster to be self-healing. When a disk or node fails, full data redundancy is quickly and automatically restored. In the case of host failure, VMs are restarted on other nodes. A larger Nutanix cluster can withstand the failure of an entire four-node enclosure (referred to as a block). NUTANIX PRISM Nutanix Enterprise Cloud incorporates management as an integral part of the solution. The Nutanix Prism management interface delivers consumer-grade simplicity for infrastruc- ture management and makes it easy to keep infrastructure up and running.