What is our primary use case?
My usual use cases for NetApp SnapMirror are primarily for business continuity reasons, allowing me to offer customers the chance to fail back services to another data center, from Milan to Rome and vice versa, or from Milan to London, or from London to Dublin, for instance. It supports very long distances since it's asynchronous, not synchronous, so any latency is allowed.
What is most valuable?
The features of NetApp SnapMirror that I have found most valuable include the possibility to stop synchronization in case of high network latency and to limit the bandwidth throttle. This means limiting bandwidth when less connectivity occurs or opening wide the throttling to use the full speed of network connectivity, which we can adjust according to our needs or the needs of our customers. All of this is done without limiting business continuity, so business continuity is always guaranteed.
NetApp SnapMirror's data replication feature supports disaster recovery strategies, as it is incorporated into the business continuity feature. SnapMirror is strictly related to snapshot features, a core part of ONTAP, so it's very flexible. It allows us to start with just two snapshots as a simple backup and then use them as a replication point or base for both backup and business continuity. There is no right or wrong answer, it just depends on how it works for specific needs.
What needs improvement?
Some areas of NetApp SnapMirror that could be improved or enhanced include synchronous SnapMirror, which needs improvement as it suffers significantly from network latency. In particular, synchronous replication with read-write on both sides in both data centers is powerful but risky since it is greatly affected by high network latency. It has been improved in the latest versions of ONTAP, specifically in 9.16.1, but Dell Technologies has made some steps forward regarding synchronous replications.
For how long have I used the solution?
I have been working with NetApp SnapMirror for 13 or 14 years, which is quite a long time, as I was working with another company and also as a NetApp partner for a limited time.
What do I think about the stability of the solution?
In terms of stability and reliability, NetApp SnapMirror is very stable regarding the common asynchronous replication. However, the synchronous replication requires a very stable network connection with very low network latency, so certain conditions must be followed.
What do I think about the scalability of the solution?
NetApp SnapMirror is very scalable, as all we need is an ONTAP cluster, starting from two nodes up to 24 nodes with NAS or 12 nodes in a SAN in an iSCSI environment, meaning that there are no limits apart from the number of controllers.
How are customer service and support?
My experience with technical support specifically for NetApp SnapMirror is that I would have expected more from them, considering that synchronous SnapMirror replication is quite a recent feature. I would rate them seven out of ten.
How would you rate customer service and support?
Which solution did I use previously and why did I switch?
I have always worked with NetApp SnapMirror since I started working as a storage engineer. I came across some technologies from Dell Technologies storage, but I do not have deep knowledge of them as I do with NetApp.
How was the initial setup?
My experience with the initial setup and deployment of NetApp SnapMirror is that it is very straightforward. It doesn't require engaging a field support engineer team, as it can be easily set up by our side, given we have some experience. Compared to MetroCluster, it doesn't require much time or expertise and can be set up in about one hour, which makes it very quick and much less time-consuming.
What other advice do I have?
I started working with NetApp SnapMirror on ONTAP 8, both ONTAP 7-Mode and the clustered Data ONTAP, which is strictly related to the ONTAP. ONTAP was not clustered in the ancient times, and SnapMirror has been used and offered to customers since only ONTAP 7-Mode was available, which works on a HA couple of storage without the possibility of clustering. I then started working with a real cluster that involves several HA couples of storages up to 24 nodes, which was quite stimulating and very scalable horizontally.
Regarding the synchronous replication feature of NetApp SnapMirror, it limits the distance between the two data centers, the source one and the destination one. It's very challenging since the two data centers need to stay in the same area. We have primarily used synchronous replication for SAN, particularly iSCSI LUNs. NetApp has introduced the possibility to write anywhere on either data center, which is a very interesting feature; however, it requires very low latency and is much more critical than asynchronous SnapMirror.
I have utilized NetApp SnapMirror's network compression, which works very efficiently as it reduces the bandwidth necessary to synchronize a replica volume to the remote data center. I can evaluate the impact of NetApp SnapMirror's network compression on bandwidth usage by monitoring the replica process, where I can see how much data is replicated as well as the network compression ratio, which is indeed a part of the whole mechanism.
NetApp SnapMirror's compatibility with NetApp products has helped my storage management efforts immensely, as it is part of ONTAP and all architectures based on ONTAP, including ONTAP 8 and 9. I can use it with all families, FAS, AFF, ASA, and AFF families.
I would assess NetApp SnapMirror's integration with cloud environments for maintaining workload performance as the maximum, as it is fully integrated with ONTAP. It is very much based on the snapshots, making it very transparent and seamless in its operation.
I have seen tangible benefits with NetApp SnapMirror, for instance, compared to MetroCluster, as it is a much cheaper way of implementing business continuity. It doesn't require a very articulated configuration, unlike MetroCluster, as we just need two common HA couples to set up a SnapMirror synchronous relationship. This allows us to offer business continuity to our customers without a fully articulated configuration, making it much cheaper.
NetApp SnapMirror is deployed on the cloud in my organization for some customers who require business continuity, particularly for asynchronous replication, which is used for backup.
On a scale of 1-10, I rate NetApp SnapMirror an 8.
Which deployment model are you using for this solution?
On-premises
If public cloud, private cloud, or hybrid cloud, which cloud provider do you use?
Other