What is our primary use case?
We are failing over approximately 250 systems. In many ways, this could impact 3,800 insurance agents across 11 states.
There are two sites: the source site and the production site. Those are failing over to another data center about 150 miles north of my location.
How has it helped my organization?
When we went from the original DR plan that we had with Double-Take to SRM, we were able to fail over in an hour and a half. We did all the storage groups in bundles, and we are like, "Wow, this is unbelievable. This is awesome." Then, we went to Zerto, and it was like, "Oh wow, we can pick and choose how we want to do this." So, Zerto provided us with a lot of value.
We went from testing in a week, e.g., we would say, "Alright, we are going to set aside Monday through Thursday to test all the apps which have been deemed 'need to be tested', and make sure for DR purposes that they are working correctly." We went from that to a day. We can do it whenever we want much easier than before. Instead of having to do it in a group, you could have it where there is scratch space and all the things that are needed, where all the changes and deltas are being cached. Now, we can do a small group of people anytime that we want, or whenever.
We haven't done it all in a day. Our plan is to have it fail over where we can get it done quickly enough in that morning, e.g., if we have all the testing, testers, and developers lined up, then they can test and we can have it done all in one day.
It has reduced staff stress. We are not big on cutting staff because we run pretty thin. We have even seen growth in the amount of staff involved in backup and DR management. There will be two leads going forward, sharing the primary duties.
What is most valuable?
The most valuable feature is the failover testing and being able to do that in a granular fashion. We can pick and choose what we want to fail over and at what time, then how quickly it fails over. We fail them over into a bubble, which means our developers and other testers can go in and do whatever they want. They are not impacting production outside of the bubble.
The reporting function is a big thing that we like. Our upper management and execs are always like, "Hey, we need to report about what you did." So, we can print out a report that is 200 or 300 pages long, and go, "Here you go." It was a little overwhelming the first time they got it. They were like, "What?" I am like, "You asked for a report. This is the report."
For the last three years, I was a secondary admin. We got into a situation where they were like, "Hey, you're the lead. You need to immediately be the lead." I was like, "Okay, alright." So, I was able to go in and create the protection groups and replication servers. We run VMware so we were able to push that out to the hosts, uninstall and decommission stuff. I was able to get that squared away within a day or two. It is very easy to use. If I can do it, anybody can do it.
The Zerto’s near-synchronous replication is very important. We used to say, "Hey, if we don't have this and if the building blew up from a gas leak, then what would we do?" Now, it is not just disaster recovery, but there are departments of insurance requirements for federal requirements going, "Hey, do you have a disaster plan in place that will successfully run? Can you provide me with those reports?" It also checks that box since we have requirements that need to meet for customer data. They need to be able to retrieve that data, either at the running site or production site. Or, in the case of a disaster, we will need to provide them with that information. So, it checks multiple boxes.
What needs improvement?
You can create a VPG and put anywhere from one to 17 servers in that group. We build them one by one. If something changes in VMware, it would be nice to be able to go in and change that VPG, having it update without messing up. When you change them now, it only applies to the copies from the points when you changed it. I wish it would purge that older data from the past. Right now, we have to build a new VPG, which is not a big deal as it is just a few screens.
For how long have I used the solution?
We have been in the Zerto world for four years, and I am the lead on Zerto now.
What do I think about the stability of the solution?
Stability-wise, I would probably give it 10 out of 10. It is very stable. If there is something not running correctly, then it is an outside factor. It is either the admin or a connection to the other site. With the dashboard, it will show you that you have this many protection groups built. Everything is an individual green square, but when there is a problem, then you will see red. It is very simple. If it has a problem, you will see something. I have not dealt with a problem where Zerto is just not working. It is usually user error or sort of outage. It is reliable.
As far as Zerto replication and DR purposes, it has not caused us any outages.
I have answered stuff for Zerto before, and they are like, "Why do you like it?" We say, "Because it works." For so long, we had stuff that didn't work for so long, and this solution works.
What do I think about the scalability of the solution?
As long as you have the license to protect the VMs, then you can scale it as big as you want.
We are currently protecting 325 VMs. We have plans to expand in the future.
How are customer service and support?
My dealings with the technical support have been top-notch. They are very good. I would rate them as 10 out of 10.
How would you rate customer service and support?
Which solution did I use previously and why did I switch?
We had Double-Take and were replicating to a site with SunGard, then we swapped. It was kind of a nightmare for us to get it working the way that we wanted. I am sure it is a great product, but the way that we needed it to work was just not working. Then, we went to VMware SRM, which worked great and went off without a hitch.
We then wanted something with a quicker recovery point objective (RPO), and that is when Zerto came in. They allowed us to failover in a granular fashion. We could pick and choose how we wanted to fail over in DR tests. That is a big part of our DR testing. Enterprises want to be able to know that they have a successful test and can run in a failed over environment, so the test is 50% of that. The other half is, “If we had to declare a disaster, where would we be?” The RPO is two to three seconds with Zerto. I have talked to people with Unitrends and several other companies who say that you can’t get an RPO that low, but that is what we have. Today, it is very fast today.
When we need to do our DR test on a specific day, Zerto has allowed us to be able to do that in granular fashion. With SRM, you had to fail a group of servers over. While that may have changed, at the time you could only do them by storage volumes. With Zerto, it didn't really matter. It has been like, “Which ones do you want to fail over? Do you want to do just your SQL servers?” This has allowed us to have a more granular approach to testing and DR testing. It ensures that we can do it in a certain way and confirms that our actual DR plan is a good plan.
We didn't have anything that worked for so long. I think Zerto kind of showed up and was in the great spot where they couldn't be any worse than what we had.
How was the initial setup?
I was not involved in the initial setup. This has been kind of thrown in my lap, and it has not been a nightmare at all.
What about the implementation team?
The prior admin hired services for updates. Going forward, I will probably do them myself.
What was our ROI?
We have seen ROI. It reduced the time for failover and failback by 90%. I am not saying that the products I mentioned earlier are bad products. They just didn't work well for what we wanted.
Zerto has had a significant impact on our RPO. It is a double-edged sword where our RTO and RPO have allowed us to almost not miss a beat. In a DR test, we are more staging and moving systems over, and this is more of a tactical approach. With some of the moves that we are making with SQL and using blue-green environments, I don't think we see a problem at all. We feel very good about it.
What's my experience with pricing, setup cost, and licensing?
We bought it through a reseller.
We are very fortunate because our budget is pretty big, and I am not making that up. Staffing may be a little thin at times, but as far as budgeting what we buy, the price for this solution has not been so outrageous that we don't buy it.
I think there is a support cost.
Which other solutions did I evaluate?
I was a big proponent of using SRM because I manage the VMware environment. Being a VMware product, I was more in their corner. So, it was mainly between SRM and Zerto. We also might have looked at Rubrik.
With other vendors that we used, we would sometimes start on the weekend, e.g., on a Saturday morning at 6:00 AM, then we would go through at least Thursday of the next week. It would be a long, arduous process. Sometimes, we would go only two days because we could never get past a single spot, then the entire test would be a failure. With Zerto, it has reduced our DR testing time drastically. It went down to where we think we can do a test in a single day. We were able to pull it off last year in two days with failover and failback tests as well as reports.
Zerto provides ease of use when building out jobs, then having them failover as you want, one by one or selecting five or six VPGs at a time. One of the big things that we do is with SQL. We want our databases online before doing any testing. There also needs to be domain controllers turned on for people to be able to log in. It is like, "Alright, we are going to fail over the domain controller." Next, they go, "Alright, we are going to fail over our SQL stuff." Before, when we had those SRM groupings, it would be a bit harder. You had to wait for everything to finish. Now, it is granular, where you can pick and hit one by one what you want. The database administrators can go in, and say, "Alright, we are online. There are three more that just came online." They are able to test it, and it just works. Having something that works was a big thing for us.
It has not replaced any of our legacy backup solutions. We use Veeam for any backups or system restores at this point. So, Zerto's role is just for DR.
We have luckily not had to use Zerto in a data recovery situation for ransomware. We have had one instance where we were in a spot like that, which was about two years ago, and we were able to restore it back with Veeam.
Until the last few cases, VMware support is some of the sorriest support that I have had.
What other advice do I have?
I would recommend Zerto because it works. You will need to do a PoC first though.
Immutable data copies are something that we are looking into. For example, if I have a recovery point of two, nine, or 10 seconds, then we get hit with some sort of ransomware attack or something like that. We would like to have immutable data that is unchanged. So, we are looking into this feature now.
I am sure it has enabled us to do DR in the cloud, but we are not a big fan of putting that stuff in the cloud. We are not a fan of putting it on somebody else's computer if we can put it on our computer. We have been very happy having a DR site approximately 150 to 200 miles north of our main site. We are kind of running it in our own hybrid cloud at the moment.
As far as testing, there are probably 70 people who test.
I would give it a nine out of 10. It has done what we wanted. We have been very satisfied with it. We are Zerto fans.
Which deployment model are you using for this solution?
Hybrid Cloud
*Disclosure: PeerSpot contacted the reviewer to collect the review and to validate authenticity. The reviewer was referred by the vendor, but the review is not subject to editing or approval by the vendor.