We use it for infrastructure monitoring.
We use it for infrastructure monitoring.
The improvement has been the ability to deploy a log monitoring system on already existing infrastructure. When I took ownership of teams where I replaced the previous monitoring solution with Centreon, the capability to stabilize and quickly fix the infrastructure monitoring - since the previous solution was not very good - helped us.
It also gives us a good overview of our system.
The collection part, due to the flexibility and scalability.
We have provided feedback to Centreon directly. The most important issue is the capability to interconnect with other systems. It already exists for some of them. For example, the Stream Connector is something we use to populate data in another system. This kind of facility for connecting should exist for all products that it makes sense to have connected to a monitoring solution.
We have never had a big issue. Sometimes there is a mistake, the wrong configuration for example. But there has been no outage caused by a bug in the software.
We haven't had any scalability issues with the monitoring solution part. But in terms of data, analysis, we have a very large amount of data, a big database. The MBI, the Monitoring Business Intelligence part is very important for that because we have limited data analysis, graphing, and reporting. With the new solution, we will have success with that. This improvement will enable a lot of functionality and features in terms of Big Data treatment.
We have direct contact with a pre-sales engineer. We work directly with him by email or by phone, it's very fast. I don't use the support gateway from Centreon.
We had a different monitoring system. Why Centreon? It's very good at agents, in collecting data. Because of some specific integrations we must do, we chose Centreon. Based on demand and our expectations, and the collectors, we made our decision.
We picked Centreon because, after an open-source comparison, and given what we want to do, and our skills and experience as well, it was the best solution for us, the most reliable for our services. We manage the unified collaboration clusters for an organization in Europe. The difficulty is that we have to maintain a complex solution so I don't want to increase the complexity of managing it. The opposite would be the best way, to effectively simplify the management of this kind of cluster. Centreon is made for monitoring but also has extra connectors to enable automation. Globally, that was a very important point for us, to open the door to tomorrow. If we stay with Centreon, we should be able to play with the data, with Centreon and with other systems.
The first setup was eight years ago and it was not very complex. We had an expert in Linux so it was not really a big issue, with ten years' experience in monitoring systems. There were no specific difficulties with it.
We have a good monitoring system and a good aggregation layer. That means we are alerted on service impacts and not when something is down. We have saved a lot of time on non-business-hours intervention.
I deal with technical parts, not with the costs.
In terms of licensing, the big question will be the ability to be flexible with the scale of design.
We had different solutions. We had Splunk for Big Data monitoring as well as Elasticsearch and Zabbix. These are at the top of the market, along with Centreon.
My advice is to make it simple. When I say that, it's in terms of templating. With Centreon we can create a lot of templates. It is very good to have something very flexible and configurable. But be careful, don't create a lot of templates that will clash with other templates because, in the end, it will be very complex to maintain. Start simple and maintain up-to-date documentation.
We use other reporting solutions to complement it, to create beautiful reports that are specifically requested by our customers. In the future, I expect we will use a diverse range of products to give us the value we need to present to our customers.
I'm a solution architect, so my main job is to provide good solutions to meet demands. When we build a design, we study which solution will make sense for the customer. As an integrator, of course, I need to be sure that any solution, for the price, will make sense for my enterprise as well.
If we compare Centreon to another open-source monitoring system, and we're talking about it as a pure monitoring system, I would rate Centreon between eight and nine out of ten. If we compare it to a Big Data system, it would be closer to seven out of ten, due to the Big Data capacity that we don't have with Centreon. Strictly on monitoring, it gets a good score but with the new technologies, what we see with Big Data and the capabilities for machine-learning and AI, etc., the latter will have a better score because they have the capability to generate a lot of metrics.