Dell RecoverPoint Reviews

• Replication over IP and Replication over FC providing CDP - Continuous Data Protection
• Multi-site Protection is a very useful feature when having more than two Datacenters

We can replicate Data between Data Centers synchronously or asynchronously. We use it to provide Disaster and Recovery copies and migrate Data between Datacenters.

Currently the product does not support cascaded replication, only "Y" replication. For example. It is possible replicate data from 3 locations and keep three copies of data Site A -> Site B and Site A -> Site C, but is not possible replicate the same data keeping three copies of same data from Site A -> Site B and Site B to Site C.

3 years

Deploys works fine in mos situations and environments. When using WAN IP Connections to connect two Recoverpoint Clusters, it requires many firewall rules, but is not a product fault or bug, just a bit complex for the firewall team.

In version 4.1 we needed to disable the new WAN Compression Feature. We opened a case with the vendor due some warning messages. EMC informed us that Recoverpoint 4.1 has a issue addressed on engineering related to new WAN Compression feature.

No, the product is very scalable. I'm able to join new appliance non disruptively and balance the replicated consistency groups into new added appliances..

Customer Service:

Very nice

Technical Support:

Very experienced support engineers. I rate it very high.

No, I've never used other solutions based on appliance integrated with Storage providing continuous data protection.

May be a bit complex due your IT environment and WAN connections, but in most situations it is a plug and play solution.

Very experienced support engineers. I rate it very well

Product is self-payable. Due data protection provided to IT environment, we can reach very secure operational levels.

Initial investment setup costs about $33K to protect three sites including 5 Storages. Day-to-day costs about $2 due to vendor maintenance contracts.

Look for new available features when product is integrated with EMC VPLEX to provide Metropoint Protection using a Three Datacenter Layout.

Main feature we use is remote (async) data replication, over IP, over a distance approx 300 kms.

• Data replication with Recoverpoint over a long distance without FC is simple and delivers a cost reduction (ethernet DWDM is cheaper than fiber)
• Our site disaster recovery plan was simplified by scripting the failover procedure.
• Offloading the storage array with replication tasks was very useful to us because in the past our experience with Mirrorview /A icm with high data change rates was sometimes problematic.

I am not familiar with improvements in Recoverpoint v4.xx, so I don't know exactly what the latest improvements are.

• EMC Recoverpoint/SE
• Software version 3.5 SP1
• Bi-directional data replication between two VNX5500
• Total amount of replicated data 25 TB.
• We've been using it now for about 3 years. (same hardware, some newer software versions)

No, together with EMC support personal we had a good deployment. The sizing was done well, the only puzzle was to set up the consistency groups in the right way and spread them out over the Recoverpoint nodes. In the end this was done by 'trial and error'. But later on we would use the commandline to monitor the loadbalancing.

No problems at all.

No, though some CG's (mostly VMWare ESX) are busier than others. Overall performance is good. Each datacenter site consists of two nodes (local cluster).

Customer Service: Very good, EMC did support us to implement it well and gave useful instructions/ lessons to help to maintain it well.Technical Support: Excellent

Yes, EMC Mirrorview /A and /S. Because of relocation of our datacenters the distance between them increased enormously because of the large distance and the availability of Ethernet only between the sites.

Initial setup was straightforward and data replication did work very quickly. The main puzzle was to seize the replication sets and consistency groups. Our host environment is rather complex. (Open VMS, Linux, AIX, Windows, ESX)

Vendor team, their expertise was excellent.

Actually I do not know but due to merging and organizational changes the solution will run for three years total.

My advice is very positive. There is a good plug-in VMWare vCenter for SRM, stretched Windows clusters, local replication feature, granular level in Point In Time Copies.

Complete out of band block replication with EMC storage arrays (such as VNX, VMAX, & VPLEX). It's seamless integration into VMware's Site Recovery Manager is very useful.

We have been able to achieve a very low RTO & RPO with near real-time replication policies & adequately sized journals for a limited recovery time requirement to bring systems online as well as a great recovery point to roll back to in the event of a disaster or outage.

The new Gen5 appliances have adopted the Unisphere management interface & it is written for Adobe Flash, which in my experience, is very slow & makes administration cumbersome at times. The Unisphere interface is missing some components that were taken out of the Gen3 Recover Management console such as the ability to refresh views & rescan splitters to pull in new storage on the fly without having to cancel a CG creation mid-flight.

Deployments with VNX & VPLEX are pretty simple as the splitter is installed on the SPs for VNX & on the directors for the VPLEX. The problem came into play when we began using RecoverPoint with VMAX. We discovered that the write splitting takes place at every single director port where a write occurs so in order to capture all writes, you have to zone the RPAs to every single FA port that has volumes mapped to them & are in use for RecoverPoint. It was a technical "gotcha" but we were able to rectify it.

Only when we discovered that the zoning requirements for VMAX were different than standard VNX deployments did we run into issues; however, we were able to quickly resolve that issue.

This solution is quite scalable. The maximum number of nodes per RP clusteris 16 so there's no issue there. The caveats associated with that are that you will need additional FC ports (by a factor of 4) on your storage fabric to support each new RPA. Also, if you are increasing the number of nodes in your cluster you are likely replicating more storage, in which case you will need adequate RecoverPoint capacity licensing to do based on the size of your work load.

Customer Service: Good at time but also questionable at times. EMC has grown to such a big enterprise they are struggling with their support sector.Technical Support: Good at time but also questionable at times. EMC has grown to such a big enterprise they are struggling with their support sector.

We were not using any other replication technology at that time but we are currently looking at Zerto as an alternative to RecoverPoint due to cost constraints & easibility of use of the Zerto product.

It was complicated learning the configuration on the VMAX array due to each director port performing the write split operation. VNX & VPLEX arrays do not operate this way so it is much easier to configure & manage RP when using VNX or VPLEX.

If through a vendor team, how would you rate their level of expertise? Implemented with direct involvement from EMC professional services.

We did not evaluate others before RecoverPoint. It was a package deal with our VNX purchases.

Make 100% sure it meets your needs & integrates properly. Otherwise you could end up paying additional licensing fees to make it work & you could also be put in a situation where the deployment is overly complicated due to the infrastructure you're using it in.