The true value of data protection is in making systems available after a failure. Data protection starts with surviving failure of server hardware components, then extends to recovering damage to a VM and further to recovering from the loss of a whole data center. A data protection strategy should address multiple levels of failure. It should offer the business certainty of the time to recover from each level and certainty of the potential data loss for each failure. SimpliVity has built data protection into their product from the lowest level.
A fundamental part of data protection is to protect against the failure of the storage media. SimpliVity uses hardware RAID adapters in each node. The physical server has at least two RAID controllers, and hard disks are protected with RAID6, so data is still protected while recovering from a single hard disk failure. Each node also has SSDs for performance, which are protected with RAID5 so they remain in operation if one SSD fails. The VM data is always stored on two SimpliVity nodes. This keeps data available even if a whole node is unavailable.
The foundation for SimpliVity’s data protection is their OmniStack Data Virtualization Platform. The The Data Virtualization Platform provides deduplicated, compressed and optimized storage for VMs and stores VM data and metadata that has been deduplicated. A VM can be backed up instantly without copying any data blocks, so a hundred VMs can be backed up just as fast as one. This instant backup of a VM means that you can perform backups as often as the business needs—if the business needs to backup an application every hour, then SimpliVity has you covered.
SimpliVity’s scope of deduplication is global. Every node in every cluster can use the same deduplicated data. A VM can be replicated to another data center simply by replicating its unique blocks—so whether a hundred VMs or two are being replicated, any block shared by the VMs is copied only once. Deduplication for replication across a WAN allows a slower and, therefore, cheaper WAN link.
Hyperconverged Infrastructure for Dummies
Learn everything you need to know about hyperconvergence - where it came from, how it solves performance issues, and more.
Learn MoreConsider patch Tuesday, when a hundred VMs all apply the same Windows updates. If you use storage that replicates changes, then you must replicate a hundred sets of changes. With SimpliVity, because only the unique blocks are replicated and less than 1% of the changes will be unique, less than 1% of the amount of data needs to be replicated. That hourly VM backup that the business needs can be replicated to another site every hour, then to a third site every day if you need it. Backups and replication are controlled by a policy set on each VM. No need to check whether a backup completed, just check that the VM is compliant with its backup policy.
Because each backup is truly a complete copy of the VM, VM restores are simply a matter of registering and powering on. Everything is complete in seconds. For Windows VMs, file level restore is supported.
Data protection and resiliency is designed into the SimpliVity platform at every level. The architecture allows multiple failures at different levels without losing VM availability. From simple disk failure all the way to catastrophic data center loss, SimpliVity has the answers.
