What is our primary use case?
We use it for application monitoring, monitoring of our business applications to provide our support teams with alerts around the problems and events that may be occurring within the apps.
We're a financial company and we monitor applications for trade entry, trade management, risk, and order flow. It's used for any application that's being used in the tradeflow stack.
How has it helped my organization?
Because we're a large organization, we've built some really great management dashboards that show the end-to-end flow of how our applications are fitting together and any alerts that are occurring on those applications at the moment. It provides a management-level view of the underlying data.
While we don't use it in a predictive mode—we are reactive—our reaction time is greatly improved using Geneos. We've gone from hours down to minutes.
It's hard to say how many issues we have detected and outages avoided using Geneos, but I definitely know that prior to Geneos we had hundreds of events a month that we were slow to react to. We may still get those events, but we are much faster at reacting to them, and that has really improved our interaction with our users. The confidence of our users has greatly improved because of our ability to react much faster.
What is most valuable?
The biggest benefit of Geneos is the fact that we can clearly see, if we have an alert, where that alert has come from. We can see the data around that alert and anything that might be relevant is also shown. We can very easily right-click and see why we've received that alert. That's the best part about it, that you've got all the data there with the alerting. It's the usability that it provides. It's not really the real-time data, you can get that everywhere, but rather the fact that it pulls all of those other pieces together.
Also, the Netprobe that is deployed on each of the application servers is very lightweight. That is another of the great benefits of Geneos.
What needs improvement?
One area where there is room for improvement is the log file. I would like to be able to do a pre-run on the log files. When you are testing log files for regular expressions, it would be good to be able to do a quick check up front on that side of things before you release that into production.
And more generally, there is room for improvement with the Netprobe agent performance and understanding when you need to deploy a second Netprobe versus a single Netprobe.
For how long have I used the solution?
I've been using ITRS Geneos for two and a half years. We're on version 4.12 but we're moving up to 5.3.
What do I think about the stability of the solution?
The stability is very good. Very rarely do we have an outage with Geneos. We do sometimes have outages with the agents, the Netprobes, usually due to the loads that we are trying to get them to do.
What do I think about the scalability of the solution?
Scalability is probably a weak spot. We find that once we get up to about 120 Netprobes we have to then run another gateway, but we can have multiple gateways running on the same server. So even though we can monitor, and we are monitoring, thousands of servers, we have thousands of gateways. One of the difficulties is tying all of that data together if you want to create some sort of summary reporting across all of that.
Our environment runs into the hundreds of applications, which is then multiple thousands of Netprobes, on the order of about 8,000 servers.
How are customer service and technical support?
Technical support for Geneos is awesome. Absolutely awesome. They are top-rank. They have an immediate chat facility off of their website. There's always someone there; really helpful, great guys.
Which solution did I use previously and why did I switch?
Prior to Geneos we used a few solutions, but they were very restrictive in what we could do. We could only do process up/down and log file monitoring, whereas Geneos has a huge range of other things that we can monitor.
How was the initial setup?
We're very thankful for that, in terms of the deployment process, we got some advice to get some understanding about how we would like to configure Geneos. You have to do some background work and upfront work before you start your deployment. Once you've done that, you can create some standard configurations. Then we deployed to all of our applications, created a standard package for how to deploy the Netprobes internally, and then we deployed the gateways onto virtual servers.
We've gone for a federated model where we have a group of individuals that are monitoring the monitor. They're monitoring Geneos and doing all the server-side configuration that is required. We then have the application support teams that have slightly less Geneos knowledge, but they set up the configuration for their individual applications.
We deployed it to over 120 applications. The initial deployment took us about six months, and then we spent another six months to a year refining that. It's a constant process thereafter of continual improving of the monitoring. Whenever we find something new, we will add in a new configuration.
The core team involved in the deployment included four people, and the extended team ran to 20 to 25 people because it's such a large deployment. When it comes to users of Geneos, front-end, it would be around 350.
What was our ROI?
It's very difficult to put a number on it, but I'm sure that the product has paid for itself.
What's my experience with pricing, setup cost, and licensing?
It is expensive. They have to look at the model around when we move to cloud and how that's going to work. The licensing cost does pay off because of the improvements in support to our business.
Which other solutions did I evaluate?
Our previous solutions and this one take roughly the same amount of time to alert. Both do so within a couple of minutes.
We had a few choices to evaluate, including using an OpenStack or open source solutions, but when we evaluated those we felt that, given the time and effort to stitch them together, it was more beneficial for us to use Geneos. The time and effort needed to get those open source systems up and running and combined—you have to use multiple open source products to get the same functionality—and the fact that there's no support and nobody to help, led us to conclude that it's much easier to use the ITRS solution.
What other advice do I have?
The configuration can be very flexible, which is great when you're trying to do something, but it means that if you're just starting out, there is more than one way to do something. Take professional advice, use the ITRS website, understand how everything fits together first, and then proceed. Try to get a consistent configuration design across all of your applications.
A lesson learned from using Geneos would be to get buy-in from the people who are the users of Geneos. Get them onboard nice and early. Get them familiar with the tool. It does take time to understand how the tool works, so invest in training.
Which deployment model are you using for this solution?
Which version of this solution are you currently using?