What is our primary use case?
DR is the primary use case, protecting our mission-critical servers throughout the company. We're spread across North America at 11-plus sites. They are all Windows environments, as far as servers go. We fit the core model well, being all Windows.
It's on-prem. We've got an HA and DR in two different locations.
How has it helped my organization?
In the old days we were manipulating jobs, changing tapes, spending time validating backups. This does everything for you. It really is a "set and forget" type of appliance. For the most part — although occasionally there is a problem — that is the way that it operates: Set it and forget it. I save five to ten hours a week.
Also, in the past, we had a backup application and tape drive which were always in close proximity to the servers that were being backed up. With Quorum, I've got it sitting in my data center, in the center of my private cloud, and I can point that to any server in our WAN, regardless of where it's located, and perform daily backups and recovery. It gives me a huge, broad reach, instead of buying backup software and backup media for remote sites. I've pulled all those into the central core and take care of it all at once. In my case, I'm backing up about 30 servers across North America without touching any of them, without looking. It just gets done.
It's helped when I've lost a couple of servers. For example, I've had a virtual server fail and I ran that server off the Quorum for a month, because it allows you to virtualize and run those VMs indefinitely off the Quorum appliance. I have also had scenarios where I've gone through the QUARK (Quorum Ultimate Automated Recovery Kit) recovery program and recovered an actual server. I haven't had a lot of failures, but I have had both types you would typically have to deal with. It has worked out fantastically for us. I've also had a scenario where the hosting company for our hosted VMs had a failure. I was able to run the VMs off Quorum until the hosting company recovered. It's not always you that fails, sometimes it's your third-party services.
In terms of the impact on our overall storage efficiency, we're not adding tape libraries, and we're not adding huge arrays of storage for archival data. The Quorum encompasses de-dupe and really minimizes the amount of footprint to store those data backups. I keep my retention at about 40 days for most servers, but in the case of the email I do 150 days, and on an email server that's quite large. That's a tremendous amount of data. The Quorum does a fantastic job of compressing that down and being as efficient as possible. An Exchange database changes by the second. When you back it up, you end up backing up the whole thing, and Quorum has a very efficient way of doing it. Any other way — oh my God — I'd have 100 terabytes of storage sitting there just for Exchange.
What is most valuable?
The "set it and forget it" approach of Quorum is huge, as far as time savings go. And the fact that the Quorum validates the data that's backed means I don't have that mystery of what's sitting on the tape. Every backup is validated. I know that I have good, recoverable data at all times.
There's virtualized recovery within the Quorum appliance, which we do use occasionally — and sometimes we'll use it for sandboxing or testing — and then there's the situation where you actually lose a physical server. On my remote sites I don't have virtual servers, I have physical servers. They've got a great recovery tool for rebuilding bare-metal hardware as well. They've got virtual recovery, file recovery, and bare-metal recovery all nailed down perfectly.
The automated testing functionality is part of the "set it and forget it" because I can look in an email that says all my VMs were backed up and tested and it takes me ten seconds. That's as opposed to having to go manipulate the system and do any kinds of validations or changes. It's stupid-simple.
In terms of the solution’s ease of use for recovering what is needed from a backup, there are two facets to it. On a day-to-day, "Suzy lost a file" type of situation — just file recovery — it's extremely fast and simple. With four or five clicks you find your file, you hit recover and, in ten seconds, it's on the server.
In a disaster situation, where you actually have to recover a full server, they have a very well-designed method. They call it the QUARK. It's the Quorum recovery procedure for doing recovery of VMs or recovery of bare metal. It's a guided GUI recovery and it works really well.
What needs improvement?
We tend to run a broad generational range of server OS's. We have everything from Windows 2003 to 2016 in production. The only thing that I've ever found fault with is that they have phased out support for 2003 Servers. In our case, we have quite a few of those in production. That's a risk we take by doing that. We can't expect the world will stop and wait for us forever. But that really is the only thing I've ever had issue with: If I lose an older one it's a little bit more difficult to recover. They can recover it, by the way. Not all is lost.
For how long have I used the solution?
We've been using OnQ for five or six years.
What do I think about the stability of the solution?
The stability is rock-solid. I've had very few problems with it and most of the problems I have had were my fault. There are not frequent updates so you are not constantly updating and changing your environment. The updates seem to be well thought out and delivered when they're needed. The system is extremely stable and has been since day one. They've added a few features over the years, modified the dashboards slightly, but overall it's absolutely stable. Like I said, you set it and forget it because you know it's going to run.
The Quorum itself, generally, is fault-proof. I might have a server that gets a little bit out of whack and which needs to be rebooted. And then, all of a sudden, the backups are working again. Nine times out of ten — and probably even much greater than that — it's your end server, your endpoint, that's causing any issues with a backup.
What do I think about the scalability of the solution?
There are two components to scalability with the OnQ appliances. There's the intended purpose of the unit, and it's sized the way you buy it. You buy it with this much compute power and that much RAM. For the intended purpose of the Quorum, if you have a DR scenario, there is a limitation by design. If you buy a box that only has XYZ memory, you can only support so many VMs until that XYZ of memory is consumed. That's one side of limitation. It's limited to what you scaled it to be able to support. If you bought it to support 15 VMs in production, directly running on the Quorum, that's what you get.
If you're talking about how hard can I run this product, when I bought it I could run 12, 13, or maybe 15 VMs on it. That's what it's sized with. I'm running much more than that. I've pushed it much farther than it was ever intended to do, and it runs fantastically. We're backing up way more servers and volumes than are offered to us to be able to do, and it does it seamlessly. No issues whatsoever.
Our company is growing rapidly through acquisitions and I would like to leverage the Quorum. I'm looking at replacing my Quorum this year with a substantially larger version of the same OnQ HA and DR appliances. My long-term ambition with this is to grow it throughout the corporation.
How are customer service and technical support?
The tech support is excellent. I know a lot of their tech support people by name. They're extremely friendly to work with, very knowledgeable. And recently, they've become a lot more proactive. It used to be that if you filed a ticket they would usually call you right back or answer on the first ring. Now I'm getting alerts from them. That's a great step forward from them.
The number-one feature of Quorum is their customer service team. It's fantastic.
Which solution did I use previously and why did I switch?
We were using Backup Exec and a tape library. Tape was slow and unreliable. There was no validation of what was recorded on the tape, and recovery on sequential data is extremely slow.
How was the initial setup?
The initial setup was very easy. It was a guided setup. The sales engineers did some of the pre-work inside Quorum before the box shipped, and they guided us through the rest.
The solution is integrated with our network. That integration process was incredibly simple. It's just another node on the network. No further configuration was needed. I just assigned an IP address and they were ready to go.
The initial deployment took days. Most of that was waiting for the initial feeds to get done and replication between the primary and secondary units. There wasn't much implementation strategy. It was almost what I would call a turnkey system: Hit ground, plug it in, and it's running. It's as close to turnkey as you can get.
What about the implementation team?
We did not need a third-party. It's very simple to use. Once you do the installation for each server, which is GUI and easily run, you tell them to activate and they're going. It's pretty straightforward. One person can do it.
What was our ROI?
There's an ROI based on the fact that we've recovered lost data and lost server environments.
There is some fuzzy-logic ROI built around RPO and RTO. In our case, we can build an ROI leveraging risk and the cost of risk, because in our company we supply assembly lines in a lean manufacturing environment. We have to supply assembly lines very quickly. We're taking orders in real-time. For us to have an outage that takes several days to be remediated would cause us to shut down assembly lines, send home union people, and have tremendous fines as a supplier. The Quorum gives us an RTO of ten minutes, in some cases. We have that risk-avoidance where we know we're never going to jeopardize our clients because we have this in place. For us, this is a huge insurance policy. The risk in shutting down an assembly line for a company like a John Deere, for example, is tremendous, for us. A scenario like that could get into the hundreds of thousands of dollars quickly. We could end up losing a contract over it.
However, we don't typically build ROI around what I just mentioned. We have risks and it's an insurance policy for us, more than an ROI. I'm not going to gain savings having it in place, I'm going to avoid risk.
What's my experience with pricing, setup cost, and licensing?
We were on the original licensing model. They've changed it over the years. Now, the licensing is based on how many nodes I want to be able to run as a virtualized recovery. In the old days, you just bought it by size. The short answer that there are really no licensing issues or costs.
The only ongoing cost is my annual support. It's typically 20 percent of what the original purchase price was, which is standard in the industry for just about everything.
Which other solutions did I evaluate?
We looked at Zenith, Unitrends, Double-Take, and Veeam.
Quorum was light years ahead of everybody else. There was backup-to-disk. Anybody can do that. Quorum's product came out of the defense industry, which gives you a feeling for how rock-solid it is. After it backs up your data, it builds the VM, and that VM is parked there, ready to start at a moment's notice to do recovery. The fact that you can fire up that VM in about three minutes' time — nobody can touch that. It's an amazing feature and they do it so well.
Every server that gets backed up every night, the data is tested, the VM is created, and it's ready to go. You know it, you've got a green light, that thing is ready to go. When it comes to recovery time objectives, RTOs, most people offer you an SLA in the two- to four-hour range or more, and the Quorum is minutes. Nobody can touch that. That's absolutely huge.
What other advice do I have?
The biggest lesson I've learned by using it is how much time I wasted on other backup methods during the previous 20 years of my career. The best thing about the Quorum is that you absolutely know that the data that you're backing up every night is good, legitimate, recoverable data. It lets you sleep at night. When you're doing mission-critical things, that's an absolute must. Data is no good if it's stored away where you can't see it, you can't validate it. Knowing that it's there and ready to go is hugely important.
When I talk to people about it, the ability of this thing to recover in seconds or minutes is amazing. The fact that my VM is built and tested to make sure it's startable, every night that backup runs, that's amazing. That's what I like about it.
A lot of people haven't heard about it. It's not one of the mainstream systems out there. It's not a McAfee-type brand. A lot of people look at Veeam as the standard and we've looked at that a couple of times over the past few years, but the RTO times are huge. It's just not an option for us. The alternative is to spend a ton of money to get something that's basically a hot failover. You can go to the VMware Site Recovery Manager and just hit the "Easy" button and failover instantly, sure. But it costs thousands and thousands and thousands of dollars to do that, and that's where I make money with the Quorum.
It really does do everything I need it to do. I'm at the 100-percent-satisfied level with it.
As for reducing expenditures because I need less hardware, it does so in the off years. I chose to do the physical hardware method. I have the OnQ server, HA, nearest to my data center, and then I have an OnQ DR server at one of my remote sites as my backup. When you're buying physical hardware you have to replace it every seven years or so. I'm going to have to replace my hardware simply because Dell was the original blackbox hardware vendor and they won't offer a warranty on the hardware due to its age. I'm going to have to do a move and replace, so there will be an expense there.
The hardware has certainly come down in cost. I spent significantly more for the original box than what the replacement boxes are going to be. The total cost of ownership is not bad because I did get a long run out of the lifetime of those. I do have the option to go into cloud. I just don't want to do that.
In terms of maintenance, I don't allow any else to touch it. It works so well and it's so well-tuned I don't want to take the risk of anyone modifying it. I've always kept it under my control. There's one throat to choke for DR. I spend 15 minutes a week looking at the Quorum. Maintenance is ridiculously small.
The solution is easily a ten out of ten. It's a great product and we're happy to endorse it and kind of spread the gospels of Quorum around.