As you navigate your digital transformation journey, you must constantly modernize your data protection architecture. For business continuity and disaster recovery, today’s organizations require the following capabilities in a data protection solution:
- Zero data loss (zero RPO), especially important for mission-critical application data, by replicating each I/O as opposed to traditional point-in-time backup methods using space-efficient snapshots.
- Zero time taken to recover data (zero RTO), for operational, compliance, and customer experience reasons. Applications must fail over to secondary storage without requiring restart or host reconfiguration—as simply as “flip-flop” failover/failback and even granular failover (at the volume level).
- A hypervisor-centric approach to taking application-consistent snapshots, being able to manage replicas, and taking advantage of the compelling economics of virtualization.
- Harnessing the power of the cloud (capability of disk-disk, disk-cloud, or disk-disk-cloud backup) to provide data protection across platforms and environments.
IT demands are changing rapidly, requiring the ability to rapidly repurpose and reconfigure data centers. Multiple or new management tools are not an option. Customers are asking for a single business continuity solution that meets all requirements—from small numbers of volumes and single applications to entire large clusters and multisite environments.
In today’s digital world, loss of vital business data can cripple a company; any downtime can have serious repercussions to the organization’s financial performance and competitive advantage. To safeguard data from loss due to natural disaster, fire, application failure, or software malfunction, you can use NetApp® SnapMirror® Synchronous (SM-S) replication software. SM-S is a volume-granular, synchronous data replication engine introduced in NetApp ONTAP® 9.5 data management software. With SM-S, you can achieve zero RPO and very low RTO across a data center, LAN, or metropolitan area network (MAN).
Here’s how SM-S replication works:
- After the synchronous protection relationship is set up between source and destination volumes, any application I/O from the host in the primary site is intercepted by the splitter. (The splitter sends the I/O to both primary and secondary storage systems over the interconnect replication network.)
- The storage system in the secondary site commits the transaction to NVRAM and immediately sends an acknowledgment to the storage system in the primary site. The primary site commits to its own NVRAM
- Both NVRAMs then get flushed to disk at the consistency point (CP) periodically.
- The primary storage system sends an acknowledgment to the host, reporting a successful operation. Both arrays process the transaction before an acknowledgment is sent to the host, so that both arrays are always in sync.
Creating an SM-S Relationship
The relationship between the volumes in primary and secondary storage systems is called a data protection relationship. Use the following procedure to create a data protection relationship.
Before you begin:
- All nodes in the primary and secondary storage systems must be on ONTAP 9.5 or later.
- SnapMirror needs to be licensed on the primary and secondary storage systems, as part of the Premium bundle.
- The primary and secondary clusters and storage virtual machines (SVMs) must be peered.
1) In ONTAP System Manager in the primary cluster, click Storage > Volumes.
2) Verify that the correct SVM is selected in the SVM drop-down list.
Select the volume and click More Actions > Protect.
3) On the Protect Volumes page, select Synchronous from the Replication drop-down list, and then select a suitable Synchronization Mode: Sync or StrictSync.
In Sync mode, if I/O to the secondary storage system does not complete, the application is allowed to continue its I/O to the primary storage system. When the error condition is corrected, automatic resynchronization takes place to resume replicating synchronously. In StrictSync mode, if the I/O to the secondary storage system does not complete, application I/O fails and synchronous replication is terminated. This termination keeps the primary and secondary volumes identical and provides zero data loss.
Optionally, you can click the intuitive Help Me Choose button, which helps you select relationships according to the requirements for RPO, disaster recovery, and backup. Specify the values, and then click Apply.
4) Specify the secondary cluster, SVM, and volume name suffix. Then click Save.
If you set Synchronization Mode to StrictSync, you will see a warning message. Select the Are You Sure You Want to Continue checkbox and click OK.
5) When the transfer is complete, verify that the SnapMirror relationship is created successfully and has a transfer status of InSync.
Notice that the primary volume is of type rw, whereas the secondary volume is of type dp.
Note: You can create only one synchronous relationship for a specific source volume; otherwise, the following warning message will appear:
Serving Data from a Secondary Storage System if the Primary Storage System Fails
When disaster causes the primary storage system to go offline, use the following procedure to start serving data from the volume in the secondary storage system. Before you can serve data from the volume to clients, you must first make the destination volume writable.
Note: Perform this task from the destination cluster.
1) Stop ongoing transfers to the secondary volume by quiescing the SnapMirror relationship:
- Click Protection > Volume Relationships.
- Highlight the SnapMirror Synchronous relationship, and then select Operations > Quiesce.
2) When the transfer status shows Quiesced, highlight the relationship and select Operations > Break.
Note: This procedure changes the type of the secondary volume to rw.
3) You must configure the volume for data access. NAS clients and SAN hosts can access the data from the destination volume until the source volume is reactivated.
- Mount the NAS volume to the namespace using the same junction path that the source volume was mounted to in the source SVM.
- Apply the appropriate access control lists (ACLs) to the NFS shares at the destination volume.
- Assign the NFS export policies to the destination volume.
- Apply the quota rules to the destination volume.
- Redirect clients to the destination volume.
- Remount the NFS shares on the clients.
- Map the LUNs in the volume to the appropriate initiator group.
- For iSCSI, create iSCSI sessions from the SAN host initiators to the SAN LIFs.
- On the SAN client, rescan disks to detect the connected LUNs.
- Attach the database files so that you can use them.
Meet Your Business Continuity Goals with SM-S
SM-S offers fast and efficient data replication: It transmits data immediately to a secondary storage system, producing exact replicas for use in any kind of failure. You have the flexibility to protect a subset of volumes in the cluster with replication between ONTAP storage systems on different platforms or models, which reduces Total Cost of Ownership (TCO). SM-S is easy to deploy because it requires no additional hardware, software, or networking capabilities, and simple deployment lowers overall cost. In addition, you can protect existing data for zero RPO. SM-S addresses the national, regulatory, and industry-mandated need for synchronous replication in industries such as finance, healthcare, or any industry that can’t afford data loss or that needs application availability. It delivers real benefits to organizations so that revenue-producing applications continue to serve business functions if a failure occurs.
To learn more about SM-S, watch videos from the SnapMirror Synchronous (SM-S) playlist.