In a recent article, my colleague George Crump discussed the need to simplify data protection (DP) and disaster recovery (DR) operations, with primary storage that is DR ready. DR ready primary storage would aid in DP and DR processes without compromising traditional primary storage performance. A good example of a comprehensive DR Ready storage system is Tintri’s all-flash and hybrid arrays.
The Tintri solutions are a scale-out storage platform, for virtualized workloads, that meets the expectations of organizations needing greater levels of availability and recoverability. Tintri offers a primary storage system with fully integrated DP and DR. It also integrates with existing DP and DR solutions like Commvault, Veeam and to the public cloud using S3 API based connector. The connector allows Tintri snapshots to be stored to clouds like AWS or to on-premises S3 compatible object storage.
Tintri offer two types of storage arrays. The T5000 series is an all-flash array, while the less expensive T800 series is a hybrid array.
The T5000 all-flash series can scale up to 308TB for 5000 VMs in a single 2 RU chassis and uses high density 3D NAND drives and multiple capacities with expansion options to provide flexibility, performance and capacity. This can scale out to 10 PB of all-flash storage in 1.5 racks, supporting up to 160,000 VMs. Tintri’s storage delivers performance isolation for every application and policy based per-VM quality of service (QoS) while ensuring low latency and high performance across the entire flash capacity.
The T800 hybrid series can scale up to 120TB for 3500 VMs in a 4 RU chassis.
All Tintri storage arrays use Tintri OS, which is a virtualization-aware file system. It offers VM-level QoS, analytics, data management, and automation. It supports vSphere, Hyper-V, RHEV, OpenStack, and XenServer hypervisors.
Management and Analytics
All Tintri operations are done through the Tintri Global Center (TGC), which allows access to all features, functions and configuration options. TCG gives access to powerful analytics capabilities that let you see all aspects of replication jobs (both current and historical), across infrastructure including host, network and storage (both local and remote). The advanced version also allows you to pool the capacity of multiple VMstore nodes and can be balanced based on workload requirements.
DP and DR Capabilities
Tintri storage software provides robust DP and DR features to help ensure the safety of the stored data while also facilitating rapid data recovery in the event of an outage or system failure. Tintri executes all of these features at the individual VM level. The DP/DR features are:
- Snapshots and clones at the VM level. This allows recovery of the full VM, a single virtual disk, or folder or file, with a single click. You can also clone a replica into a live VM. This can facilitate and simplify DR testing.
- Asynchronous replication: one-to-one, many-to-one, one-to-many, allowing up to four different destinations. Policies are applied at the VM or service group level, sending only changed data that is deduplicated and compressed.
- Synchronous replication: Zero Recovery Point Objective (RPO) and near zero Recovery Time Objective (RTO) ~30 seconds with one-click failover at the submount level.
- Snap recovery and File Level Recovery (FLR)
- VMware SRM integration
- RESTful API, PowerShell and vRO plug-in simplify and automate workflows
The Tintri OS provides great flexibility allowing asynchronous, synchronous and non-replicating VMs on the same array as well as supporting bidirectional replication. Replication can take place between different types of Tintri arrays, both all-flash and hybrid. Many other solutions require the exact same type of array for both source and target. Tintri does not have this limitation. The ability to use less expensive target arrays can significantly reduce costs at the DR site.
Tintri has a powerful primary storage solution that meets the DR ready standard. It provides functions and controls at the individual VM level rather than at the LUN level. This allows for selective replication of VMs that require it rather than all the VMs on a given LUN. Additionally, in a recovery situation, you can recover selected VMs without the need to wait for the recovery of the entire LUN or volume. It also has excellent DR features that greatly simplify management and recovery operations so you don’t need a DP/DR specialist to get it setup and manage it.
Sponsored by Tintri