NetApp SnapMirror Reviews

We are using the solution to backup our two data centers between our location site and our data center in Milan, and, in both directions: from Grassobio to Milan and from Milan to Grassobio. From the customer's side, we are storing all of our customer's data as an application in our cluster on our data center.

There are a lot of features that are excellent. The data compression during the transfer, for example, is good. The replication is great. The durability of the solution and the speed of the data moving from one site to another is impressive. 

At the moment, we are not using the synchronous mirror, but we can implement the synchronous mirror. The only thing is that it needs another license. This is the same issue we have with business continuity. It needs another license to be implemented. You need too many separate licenses for the features. They should add some of the aspects of the premium bundle into the basic bundle. Sometimes our customers choose other replication services because of the initial cost of the licenses.

The solution is very stable.

The solution is very flexible. We can move the destination from one site to another, or from one cluster to another without cascading the relationship, so transitioning is quite easy. 

Technical support is quite good. For example, I've had an issue in the past, and within a few minutes, with a chat, the issue was solved.

I did not previously use another solution.

Previous earlier versions of the solution had quite complex setups. Now, you can perform the cluster peer SQL replication or from the GUI so it's become quite easy and quite quick. Once you have stated the replication volume, etc., and the clusters are aligned, it takes a minute to implement. Of course, that's once your network requirements are specified.

We handled the implementation ourselves.

We are running multiple versions of the solution, including 9.3, 9.5 and 9.6 versions, and, sometimes, we are migrating from 8.2 version of 7-Mode to 9.3 or 9.5 version of NetApp Cluster Telnet.

We're a NetApp partner. 

We use a private cloud deployment model ourselves and we are also a service provider for our customers.

It's quite stable, it's not that easy to implement, but it's stable, and it's reliable, and all the data that's on the destination are read-only so there is also extra security.

I would advise it would be a good idea to implement it with VMware. It would make good base for disaster recovery.  

I would rate the solution ten out of ten. It's really reliable.

NAS Filer with quick recovery features within same site or across DR site. Being a unified storage, it is used as a Block (SAN) storage as well with pretty decent performance.

The most valuable feature is the snap shot.

It is not as efficient as it needs to be. It doesn't compress the data really well when it is forwarded to a secondary site. The download performance could be better.

Our company is Oman, and there is a ticketing problem here. If there are issues, we have to go to the partner who did the implementation. We need to contact the partner, then the partner has to get the log, and then they will raise the ticket on our behalf. After that, the tech support team contacts the partner and then it gets to us. We would like to get support ourselves or the ability to directly log the ticket or even call without having to go through the partner. We want to report the problem to tech support directly and then take it from there.

In terms of improvements for the solution, I think they already have a different product, AFF, where you can get much better performance and more combination as well as data security and data savings, if they have a different protocol already, which they are lacking now.

We didn't have any problems with stability during the installation.

The solution is scalable. The upgrade is really smooth. It doesn't require any content. Once it's done, it's ready to go. We can scale out as little as we want. We have five servers right now and 900 users.

With technical support, there are multiple levels. We are not very happy with Level One and Two support from our region. When it comes to escalation support, when there is a huge problem, it goes through the US or Canada. Those support techs are really good.

We did not previously use a different solution. It was entirely a different use case. Previously, we were using a Windows-based file server. We didn't have the high availability and replication to secondary site, also snap shot's capability, which is what we were looking for. 

The initial setup was straightforward. It was pretty easy. Most of the items were available from the GUI end.

We are using the private cloud deployment model.

I would say that if people are using this product they should make use of the snap shortcut. It would be very useful to them. If they can afford it, they should get the second box to put at a different site and replicate it for maximum protection of their data.

I would rate the solution seven out of ten.

It's a simple and efficient tool to build a disaster recovery solution that involves many applications such as Oracle, SQL, VMware, and Hyper-V.

We have a VMware environment and many NFS exports to Linux servers which are easy to replicate. If we lose the main environment, we can restore all the applications to the recovery site to the appointed LUNs, and then the NFS servers to another server using storage from the DR site.

With VMware, this is very simple because you replicate the identical volume where it is stored in your datastore. I think that recovery for VMware is too simple and you don't need any tool to support this procedure, only directing the LUNs to another ESXi and executing the procedure inside VMware to mount datastore.

In the Oracle environment, the procedure to restore the database on the recovery site is simple as well.

I think that NetApp needs to improve the performance of SnapMirror with controller volume access. This is because when you have many SnapMirror jobs running, the controllers give priority to accessing volumes and transfers remain slow. It is very difficult to understand how to change this in 7-Mode.

I have used this solution by more than 8 years.

If you have to replicate a database environment, you have to put your database into backup mode before starting the replication. If you don't do this step the data will stay corrupted when you need to use replica on site the DR.

No issue if you follow best practices and understant your replicated application/solution...

10

10

Yes i have used solution from HDS and Fujitsu and the company accquire Netapp and so their dont back more .

The initial setup is too simple because Netapp integrated the configuration snapmirror on the oncommand system manager.

I have implemented this solution since i started my working with Netapp solutions

You must understand your environment to implement any replicated solution, talk to business area before implemented and design your Disastery Recovery .

It provides block-based replication of data from the primary system to the disaster recovery systems. It’s fast and doesn’t take a lot of bandwidth, and it transfers everything including data and snapshots. Also, the clustered Data ONTAP is valuable.

It's very robust, very fast, and very easy to set up.

On very large data sets with a lot of changes, the interval between mirrors can be rather long.

I've used it for 10 years.

It has been 100% stable.

I can’t say how far it goes, but I haven’t run into problems yet. It’s very scalable, although on very large data sets with a lot of changes, the interval between mirrors can be rather long.

Technical support is very good.

No previous solution was used.

Initial setup was very easy as it’s part of our Data ONTAP system, so we just select our source and destination volumes, click initialize, and that’s it.

Keep in mind that is an asynchronous replication solution, so the destination will always lag behind compared to the source. If you can live with that, then it’s a perfect solution.

Its ability to transfer data retaining its complete integrity, access, permission etc.

We have migrated data from old and expiring Netapp hardware using Snapmirror - thereby allowing us to decomm the old hardware and save power, maintainence, cost etc.

None for now.

8 years.

None.

None.

Excellent. None of the users whose data has been migrated using Snapmirrot have ever complained back to us.

10/10, just like any Netapp support.

No - this is the best tool from Netapp and we never thought of any other tool. In fact, all additional features like Netapp Datamotion for vfilers use SnapMirror as their backend.

Straight forward - infact just 3-4 commands can effectively deliver the power of Snapmirror.

Obviously in-house.

100%.

No idea. I am a technical guy and dont get into finances.

None. This technology has been in place for last 10 years as I know and I have personally worked 8+ years on it.

Administration of Netapp Snapmirror and Netapp Storage through CLI or Oncommand manager is the best.

Creation of any size of storage LUN and replication of the same with Snapmirror environment is most valuable.

I can not say much, but obviously Netapp came out with different types of GUI based management tools for Performance management and Replication Management which has made acceptability of Netapp environment better than earlier.

I must say that Netapp will have to come out with Better performance and Faster Data replication technologies.

We have been using Netapp Storage and Snapmirror for 4 Years for Banking Domain applications, I have also seen that many clients using Netapp Snapmirror for Messaging as well.

About about a year now we have faced random issues related power supply or auto failover failback issue which was related to bugs in Firmware.

No, it's very good.

I would rate is with 3 as India Support which I am experiencing is not that good when compared to EMC.

I would also rate is with 3.

It is straightforward in terms of LUNs,NFS and CIFS with Snapmirror at DR.

Vendor Team - Yes Experience was good

I am not aware of it as I am the administrator, Client used to manage costs.

Of course. We checked for EMC's VNX and Hitachi's HNAS.

I find the most valuable features to be the management suites, integration into VMware with the Virtual Storage Console, and the capabilities it offers toward disaster recovery such as Snapshots, Snapmirror, and Snapvault.

A NetApp storage solution allows a company to architect a disaster recovery plan that protects all company data. For instance, you have a production NetApp. All company CIFS data is located there. All VMware NFS volumes are located there. All server LUNs are located there. In essence, the livelihood of the company is served up from a central NetApp storage system because all company data is there. This production storage system offers local snapshots for easy restore capabilities. You can then Snapmirror all of this data to a separate NetApp storage unit in a different hosting facility. You can also implement Snapvault. If disk-based recovery is enough you can stop there. Or if you prefer, one can implement tape-based backup via NDMP against the DR NetApp SAN then ship tapes offsite to Iron Mountain or somewhere else. This essentially gives you local Snapshot to production, Snapmirror to DR SAN and offsite tapes. This is a simple and brief description but should provide an outline to its benefit to an organization as it relates to disaster recovery or a backup solution. There are a number of different capabilities a NetApp filer provides that is a clear benefit to any organization.

This has to do with Snapmirror to an unlike NetApp unit and disks but the process is called Deswizzling. This process can get out of control sometimes and cause excessive disk utilization. You really have to be careful with Snapmirror planning as it relates to how often synchronizations are completed especially on large volumes. The other improvement area could be support for replication from Windows file servers. There is the NetApp OSSV solution but you are version locked with ONTAP. You cannot run cluster mode on 8.2.x and still support OSSV. There is an appliance but that can get a bit pricey. There are open source solutions if you look hard enough and put some effort into implementation.

I have used NetApp storage products for the past ten years.

With large implementations there are always hiccups but nothing serious that can't be fixed with minimal effort.

Initially one can experience issues with stability if the NetApp unit is over utilized. This can happen. I have seen a NetApp unit with sustained 100% disk utilization over a period of a few days. This causes real issues. This is why I say it is imperative to really plan the deployment of disk storage systems. You cannot overload the system with production access, Snapmirroring, Snapshots, NDMP, and such. Definitely give your plan due diligence and design your implementation accordingly.

NetApp systems are very scalable especially with cluster mode ONTAP.

I do not interact much with NetApp customer service. That is the job of my primary purchasing vendor. However, the few times that I have it was great service.

NetApp's automated support is great. If the system encounters an issue it will fire off an auto support to NetApp. On the rare occasion that I have actually had to call NetApp support they were very responsive and actively worked to a resolution of my issue. Overall, excellent technical support.

It is not to say I ONLY use NetApp. That is simply not true. I just prefer NetApp. I have implemented storage solutions from pretty much all vendors. It mostly comes down to a business decision of cost vs. benefit. NetApp is not cheap by any stretch but you do get what you pay for.

The initial setup of racking, wiring, assigning shelves, IP addressing, and even array and aggregate creation is fairly simple. As you integrate systems and data onto the storage solution it can get a bit complex I think. I am so used to it now it is like second nature but I can see it being complicated for some who are not used to a particular solution or vendor. Examples of complexity could be implementing high availability, fiber channel, iSCSI, NFS, LUNs, flash cache, snap schedules, permissions, Qtrees, Quotas, NDMP, CIFS, etc. all on the same production filers.

This really depends on the size and complexity of the implementation but typically it is either in-house completely or a combination of in-house and vendor support. I have never relied on just a vendor implementation. The primary vendor I use is very knowledgeable and responsive to my needs.

It is sometimes difficult to quantify a return on investment when it comes to certain technology especially when it does not technically make the company money. However, with a NetApp storage solution, if designed and implemented for a specific purpose, it can provide some real numbers for a return on investment. If a particular solution was built to provide high availability, backup capabilities, disaster recovery capabilities and faster access to data then one can clearly calculate an ROI.

I cannot provide specific numbers but depending on the size of the implementation it can run well into the $100K+ range with recurring support contract costs. With power and cooling costs it can also accrue costs quickly. Again, certain solutions are cheap while others are not. It is really a business decision as to which is selected.

I evaluate different vendors regularly to keep up with new technology from various vendors. I have recently evaluated solutions from EMC and Dell.

NetApp is a solid solution for SAN needs. It is best to really do your homework and find the solution that is appropriate for the organization and allow for growth based on company historical data. You don't want to just purchase the top model solution just to say you did but at the same time you don't want to under engineer a solution either. Take your time and design what's best for the company.