What is our primary use case?
As a bank we have a lot of retail branches, and we especially rely on Aternity for helping us do fault domain isolation across our infrastructure and in the end-user space. We can understand relative performance between different remote locations, and we can understand, within a user profile, when there are hardware issues and when there may be software issues. We use it in our corporate offices as well, but we really see the focus being around when a branch user is having a problem.
We're not as mature as some organizations so that we don't have a full, proactive reporting and alerting built through Aternity yet, but that's on our agenda for the near-term, in the next three to six months.
We deployed it in our own AWS space. It's not on-prem, but it's also not SaaS.
How has it helped my organization?
When we converted Windows 7 to Windows 10, we were able to isolate some issues. Aternity pointed out that there needed to be changes in the VDI. We needed more memory to be allocated. It wasn't necessarily clear just from the specs from Microsoft, but it became clear as we migrated people over, with a before-and-after view within Aternity.
When employees complain of trouble with applications or devices, Aternity enables us to see exactly what they see as they engage with apps. That allows us to focus our troubleshooting. Fault domain isolation is the difficult problem. Knowing where the problem is 75 percent of fixing the problem, or even more than that. Aternity helps us know where the problem is. We can compare different branches, we can compare different users, and we can compare different applications to help us determine what the common factors are.
What is most valuable?
The most valuable features are the ability to
- separate machine issues from software issues
- build custom monitoring of our own homegrown or non-standard applications.
As a financial institution, we have a lot of applications that are either written internally or bought from a vendor and customized for us. Having a tool that lets us monitor specific transactions in those applications allows us to focus on the transactions that are important to the business. We find it valuable to be able to see what's going on with the hardware and look at standard applications like Outlook or Teams or Office applications. Those provide a comparison point and let us separate out hardware versus software issues.
The custom monitoring is where we really do see a lot of value.
What needs improvement?
We don't feel that we get the back-end transaction details from Aternity. We have other tools that do that.
Also, there is room for improvement in the granularity of the alerting and reporting. We would like to be able to alert on a defined set of users for a given application, for example, that all users in this group who are using this application are seeing low performance. And we would like it to provide comparisons of that to other users in a similar group that are not experiencing the issue. We would like the ability to alert and report on those types of specifics. I don't necessarily know what all the parameters are that I might want to use to slice that data, but our experience has been that within Aternity it's not always as granular as it needs to be.
Version 11, with the Tableau reporting, offers some promise there. We're only a couple of weeks into Version 11, so we haven't fully implemented it. But that's something we're looking to improve with our new version, moving forward.
The other place for improvement, as an on-prem, non-SaaS customer, is that the system administration and management in Aternity are very difficult. They've even told me that most of their support calls come in due to configuration and system administration on their on-prem. Their on-prem solution is not easy to use. I know it's not their focus, but for now they still have us and a lot of other customers using it, and they could improve that, rather than forcing wholesale, brand-new builds.
For how long have I used the solution?
We've had Aternity for six years.
What do I think about the stability of the solution?
The stability has been good. When we were running version 9, we did not have a lot of problems. We've run into a few applications that were affected by the agent so that we had to not use the agent on some of our very specific, custom-built apps. The Aternity agent somehow interacted with them to the point where the application did not work. But stability-wise, in general, nothing has changed.
What do I think about the scalability of the solution?
Their design is pretty scalable from what we've seen. Before I was involved with the product, people did take it from just a couple of thousand agents up to 10,000, and now we're over 20,000 agents, without too much trouble. It does scale. I've talked to other companies that have hundreds of thousands of agents.
We do not have all our business-critical applications in there. It's also not just a few. We were waiting because we just upgraded to Version 11. We are looking to now go more broadly into other applications. Certainly, the most critical applications are in there.
We have plans to increase our usage. We have a mandate to start using it more for proactive monitoring and to increase the footprint, the number of applications, that we're looking at.
How are customer service and technical support?
Aternity's technical support is average. We had to push to get the right people and resources engaged from the back-end technical. I found that a lot of the support required us to wait for an email response. We've pushed our account team and they did respond and help with that somewhat, but in general I've seen better and I've seen worse than the Aternity support, in the tech world.
Which solution did I use previously and why did I switch?
Previous to this, there wasn't really a tool that gave visibility into the end-user device experience at this level. We had related solutions from Dynatrace that would look at the back-end system performance and the front-end user experience as users connected to the servers in the data center. But they didn't look at what was happening on the desktop and how the end-user really perceived that webpage loading or that Outlook item coming in.
How was the initial setup?
The initial setup was complex. There are multiple servers involved in the management system and getting those servers to interact properly — getting them configured so that the management system, the aggregation servers, and the database all communicated properly, all shared certificates properly, and had the proper certificates installed for the API — all of those pieces were difficult. There was a lot of stuff that was not straightforward in our implementation.
Our upgrade from version 9 to version 11 took three months to get the new servers built and configured correctly, tested, load balancers built, etc. That was with Aternity support, so it was not a straightforward implementation.
In terms of an implementation plan, going to version 11 we built a development environment in AWS, completely separate from our existing version 9 production environment. We got that working and then replicated it into production and then deployed part of the solution alongside the current version 9 before we finally upgraded the full system to version 11.
Internally, on our admin side, there are three IT folks who work on Aternity.
What was our ROI?
What I'm spending versus what I'm getting is a little high, especially as I explore the possibility of moving into their SaaS solution. But I think we have had return on our investment. We had some struggles under the older version, struggles that version 11 seems to be fixing. If we get to the place where we are proactively alerting and where we're giving better reporting, both of which are available in the new version, then we'll absolutely be getting return on our investment from our on-prem.
What's my experience with pricing, setup cost, and licensing?
In my opinion they are asking a lot for their SaaS solution, but I also know that that's the direction they're going. They seem like they're on the high side for what they're providing, but we're not fully implemented. We've got some room for growth. As we grow into using Aternity more, I would hope that we'll be able to do that with costs staying flat. Then it would become more of a return on investment.
Their pricing is a little high. Their pricing model has changed from the old style — and all companies are doing this — the older perpetual license plus maintenance, to more of a subscription-based service. They're pricing their subscription a little high right now.
The current, on-prem solution is probably a fair price. I need to get more value out of it, so that's where I hesitate a little bit. But especially in the SaaS world, when I looked at some of the pricing, I was a bit taken aback.
Which other solutions did I evaluate?
At the time, we did not look at other solutions. I wasn't managing the team that runs Aternity at the time Aternity was chosen. I don't know for sure what else they looked at. We have looked briefly at other solutions in the past, after having already had Aternity in place, and have not chosen to take it out, at least not yet.
What other advice do I have?
My advice would be to push the support people to help you. Engage the vendor early in the process, via Pro Services or via the support, to help with the implementation. Aternity support requires you to press a little bit to get what you want. If you want to get support, you have to engage them strongly and be very assertive.
Have a solid list of objectives for what applications and what activities you want to have monitored. It's easy to get lost in "Let's look at everything" without understanding what your key, business-critical functions are. Have a top-10, top-20, top-50 list of activities and attack them that way. That's been a bit of a weakness in our implementation.
The fact that other products may provide deeper visibility into device performance does not concern us. We've had very few cases, to date, that have required any deeper level of device performance metrics.
Right now I would rate Aternity at about a seven out of 10, and with the potential to go right up to an eight-and-a-half or nine if we get our version 11 implementation completed the way we're planning.
Which deployment model are you using for this solution?