Let’s start this post starting from an assumption: we have a bug in a company application. The first objective is to detect that something is not working correctly, then determine the origin and then solve it to restore correct operation as soon as possible. It can be a bug, a server crash, a network failure, a hardware problem, etc. We do not know. We also do not know what the origin of the incident was.
Arnaud Delanoue, Big Data Architect at PUE, explains how the consolidation of logs and metrics with Datadog can help us in the analysis and detection of problems in an efficient and agile way. And it is that Datadog allows us to unify the metrics of all the components to be able to monitor it in a more optimal way. Don’t miss Arnaud’s talk in the following video:
Investigating possible errors
In large companies, what usually happens when doing research is that each department has its role and each team works with its own independent monitoring: so we have the application support team, the network support team, the communications team, the base team of data etc.
From PUE we recommend and bet on the construction of a unified datalake of logs and metrics that promotes collaborative and efficient work. The objective of this consolidation of logs and metrics is have a single source of truth in order to analyze the problem that the application has. If we unify the metrics of all the components, monitoring will be more efficient.
In conclusion, if we want track interactions between separate components it is essential log consolidation as offered by Datadog.
Know the pillars of Datadog
The three main pillars of Datadog are traces, metrics and logs.
The traces will allow us to access the information in a unified way. We see the full path a user takes, for example, from clicking a link on a web page to retrieving or reading from a database. And vice versa.
Below, we summarize the most relevant advantages and functionalities of the platform:
- It is a very easy to implement SaaS platform. To explain it in a simple way, we will say that it requires installing an agent in each of the components, whether in cloud providers, in Kubernetes, in Docker, in Windows, etc. to start monitoring the systems.
- Offers dashboard that show different cells with the metrics, traces and logs. In this way we have a total perspective and, as we said before, a single source of truth. We can also access animated dashboards in which we see graphs that allow us to select points and directly access the logs related to that metric.
- It has more than 400 integrations, which means that if we have tools like Java, Kafka, Chef, Couchbase, among many others, it will be very easy to start monitoring them. In addition, this monitoring system allows us to have a full stack view of all our deployments. In other words, on a single platform we can view the key parameters of our infrastructure.
We accompany companies that want to undertake a digital transformation orienting to Big Data and Cloud through innovative technologies and solutions that seek to increase performance, efficiency, agility and results.
At PUE, as experts in the integration of Big Data, Cloud Computing, Microservices and NoSQL technologies, we offer you a 24×7 support service, preventive and proactive. This way of working allows us to reduce the level of incidents to less than 0.5%.
What’s more, PUEWhat Datadog Managed Services Providerimplements the platform and directly manages customer environments in addition to selling licenses within the Big Data Full Stack.
Links of interest:
Datadog Fullstack Extreme Monitoring
PUE Sessions, the event on Big Data and Cloud trends closes its annual edition with successful participation
firstname.lastname@example.org for professional services in Big Data and Cloud technologies.
email@example.com for official training in Big Data and Cloud.
firstname.lastname@example.org for official certification in Big Data and Cloud.