Every year at VMworld I have conversations that broaden my understanding and appreciation for new products on the market. This year was no exception as I had the opportunity to talk at length with Fidel Michieli, a System Architect at a SaaS provider, who shared his experiences with me about his challenges with backup and recovery and how he came to choose Cohesity. In this first installment in my interview series with Fidel, he shared the challenges that his company was facing with his existing backup configuration as well as the struggles that he had in identifying a backup solution that scaled to meet his dynamically changing and growing environment.
Jerome: Fidel, thanks for taking time out of schedule here at VMworld to meet and talk with me about how you came to choose Cohesity for your environment. To begin, please tell me about your role at your company.
Fidel: I work as a system architect at a software-as-a-service (SaaS) provider that pursues a very innovative, agile course of development which is very good at adopting new technology and trends.
My job is on the corporate infrastructure side. I do not work with the software delivery to our customers. Our software is more of a cookie cutter environment. It is very scalable but it is restricted to our application stack. I work on the corporate side where we have all the connectivity, email, financial, and other applications that the enterprise needs, including some of our customers’ applications.
I am responsible for choosing and deploying the technology and the strategy to get us to where we need to go and try to develop this division and the strategy to see where things are going. We used Veritas NetBackup and Dell deduplication appliances for backups. Using these solutions, we were constrained as they did not scale to match the demands of our business that grows at the rate we were going.
One of the biggest things that we have to worry about is scale. Often by the time we architect and set up a new solution, we always end up short. If we do not scale, we die.
We were at a crossroads with our previous strategy where we did not scale. It was very expensive to grow and manage. The criticality of the restore is huge and we had horrible restore times. We had a tape strategy. The tape guy came once a week. You could ask for a tape and it would come the next time he stopped by so we would potentially wait six days for the tape to get there. Then you had to move the data, get it off of tape, and convert it to a disk format. Our recovery SLAs were horrible.
I was tasked with finding a new solution. For back-end storage, we looked at Data Domain as we were an EMC shop. For backup software, we looked at Gartner and their magic quadrant and we chose the first three. With EMC (now Dell Technologies) we saw what the ecosystem looked like 12 years ago. A bunch of acquisitions integrated into one solution. It does not get one out of the silo scaling. There were some efficiencies but, honestly, we were not impressed with the price.
Jerome: Did you find EMC expensive for what it offered?
Fidel: Yes. It was ridiculously expensive. We also looked at Commvault and just with the first quote we realized this is way too complicated. We are a smaller organization, so we do not have people dedicated to jobs. Commvault quoted us 30 days for implementation engineers. We would have a guy from Commvault in our office for 30 days implementing and migrating jobs. That speaks about the complexity about what we are doing and it speaks to how, when their implementation engineer leaves, who is going to take on these responsibilities and how long is that going to take.
We decided that we should find a more sensible approach.
Jerome: How virtualized is your environment?
Fidel: 98 percent. All VMware. This led us to look at a virtual machine backup solution. We had heard very good things about this product but the only problem we had was the back-end storage. How do we tackle the back-end storage? My background is on the storage side so I started looking at solutions like Swift, which is an open source object-based storage as well as ScaleIO. Yet when we evaluated this virtual machine backup solution using this storage, we were not impressed with it.
Jerome: Why was that? Those solutions are specifically tailored for backup of virtual machines.
Fidel: To be very honest, NetBackup performed better which I did not expect. I was very invested in the virtual machine backup solution. We did a full analysis on times and similar testing using different back ends. We found that the virtual machine backup software was up to 37 percent slower and more expensive because of its licensing model so it was not going to work for us.
Jerome: What did you decide to do at that point?
Fidel: We talked with our SHI International representative. We explained that we experienced a very high rate of change and that we needed to invest in a solution that in 2-3 years could be supporting an environment that may look radically different than today. Further, we did not want to delay deploying it because we were concerned how competitive we would be. If we delayed, the impact could be huge.
He recommended Cohesity. We recognized that it was obviously scale-out. One of the things that I particularly really liked about its scale-out architecture is that since you originate all of your data copies from the storage, you can have multiple streams from all your nodes. In this way, you are not only scale-out on capacity, but also performance and the amount of data streams that you can have.
In part 2 of this interview series Fidel shares how he gained a comfort level with Cohesity prior to rolling it out enterprise-wide in his organization.
In part 3 of this interview series, Fidel shares some of the challenges he encountered while rolling out Cohesity and the steps that Cohesity took to address them.
In the fourth and final installment of this interview series, Fidel describes how he leverages Cohesity’s backup appliance for both VM protection and as a deduplicating backup target for his NetBackup backup software.