In my blog post on the new features of clustered Data ONTAP 8.2, I highlighted three key benefits: nondisruptive operations, proven efficiency and seamless scalability. This blog explores nondisruptive operations in more detail. For addition information on nondisruptive operations, see older blog.
Growing competitive pressure and 24-hour business cycles require that your mission-critical business processes and data be accessible around the clock. As much as 90% of downtime results from planned or predictable events such as software and firmware upgrades, replacing components, maintenance, adding capacity, and so on. With IT now an integral part of your business operations, the impact of downtime goes beyond dollars or productivity lost. Your company’s brand might be at stake. Clustered Data ONTAP eliminates sources of downtime and protects your critical data against disaster so important business processes stay online.
All clustered Data ONTAP configurations (with the exception of single-node clusters) are designed to be fully resilient to component and software failures. By making it possible to perform maintenance as it’s needed—rather than deferring it to rare planned downtime windows—clustered Data ONTAP further reduces the risk of unplanned downtime. By using NetApp AutoSupport tools to identify potential issues and performing maintenance as needed, you can reduce the number of priority-1 support cases by 80%.
Figure 1) Nondisruptive operations protect your infrastructure against planned and unplanned downtime.
A Better Approach to Maintenance and Lifecycle Operations
In the traditional storage world, workloads are assigned to inflexible physical assets. When those physical assets need maintenance, IT has to schedule downtime not only for that asset but for all the workloads, applications, and users that rely on it.
Clustered Data ONTAP offers a unique capability to support nondisruptive maintenance operations and lifecycle operations such as load balancing, capacity expansion, and technology refresh so you can perform critical tasks without interrupting your business. The ability to dynamically assign, promote, and retire storage resources lets you improve service levels over the lifecycle of an application.
Because clustered Data ONTAP can transparently migrate active data and network connections anywhere within a cluster, data can be redistributed any time and for any reason. These activities are transparent and nondisruptive to NAS and SAN hosts, allowing the storage infrastructure to continue serving data during operations such as rebalancing capacity usage, optimizing changing performance requirements, or isolating one or more controllers or storage components for upgrade.
If you consider the typical consumption pattern for storage, it becomes immediately clear why this capability is so important. In most data centers, you buy a storage system and, when that nears its limits, you add another one. Because migrating data between storage systems has traditionally been such a difficult and disruptive task, old data tends to stay where it is while new data gets written to the new storage system. Because new data is usually more active, the storage workload shifts to the new storage system, which ends up busy—perhaps even overloaded—while the old storage system becomes underutilized from a performance standpoint.
With clustered Data ONTAP, you can easily redistribute storage workloads across the storage nodes in your cluster so these types of problems go away. When you add new nodes, you simply move new and existing storage workloads to those nodes to balance the load. A cluster can contain several types of media—high-capacity HDDs, high-performance HDDs, and SSDs—so you can provide multiple tiers of storage in a single cluster and move datasets to the storage tier that best meets your needs—even when your needs change on a daily basis. Because you can move data dynamically to the most appropriate tier of storage, your storage costs are lower than under the old paradigm, in which high-performance storage tends to be overprovisioned and old and/or cold data tends to stay there indefinitely.
With the latest software release, clustered Data ONTAP 8.2, you can move data volumes even while replication or other activities are taking place, and you can perform controller upgrades without moving data volumes off the controller. That makes these operations even faster and easier to accomplish. Support for SMB 3.0 provides greater availability for Microsoft Windows Hyper-V environments as well. Earlier versions of the Microsoft SMB protocol were resilient to volume moves but failover events required clients to reconnect. A number of additional features further enhance Windows support, including:
- ODX, to offload data transfers from hosts to NetApp storage
- Auto Location, a capability unique to NetApp that enables clients to have the most direct path to storage
- BranchCache, to allow Windows clients to cache data locally to increase performance, particularly over WAN connections
- FPolicy, for quota management and file auditing
- File access auditing, to support monitoring, evidence, compliance, and recovery needs
Examples of nondisruptive operations.
Life Cycle Operations
- Rebalance controller performance, capacity and/or disk performance and utilization.
- Add storage controllers or disk shelves
- Add hardware to controllers
- upgrade storage controllers, disk shelves, cluster switches
- Upgrade storage software
- Upgrade system, disk, switch, firmware
- Replace failed controller or component within a controller such as a NIC, HBA or failed storage components such as cables, drives, I/O modules
Eliminate the Pain of Technology Refresh
When it comes time to replace your storage hardware, the process has always been disruptive, time consuming, and expensive. In fact, a recent study suggests that migrating data from an old storage array to a new one takes about five months to accomplish on average and adds almost 50% to the cost of owning an array.
Clustered Data ONTAP lets you accomplish full hardware refreshes quickly and easily without taking data offline, thereby avoiding these hidden costs.
Simplify Data Protection
To further protect your operations, clustered Data ONTAP provides integrated data protection (IDP) technologies with near-instant backup and recovery, replication for disaster recovery, and best-in-class integration with enterprise backup vendors and leading applications. IDP extends to include integrated and unified disk-to-disk backup and DR in a single process for VMware and Microsoft virtualization.
Clustered Data ONTAP 8.2 includes support for SnapVault disk-to-disk backup at the volume level, allowing you to replicate data to secondary storage for online backup and recovery. Because SnapVault only replicates changed blocks and preserves compression and deduplication savings it uses network bandwidth and secondary storage very efficiently.