Is Elasticsearch stack capable of distribute network monitoring?
Hi @musialny - thanks for your interest in the Elastic Stack. It's a broad question, but given that, I would say the answer is definitely yes. Logging and monitoring are very mature stories, and Elastic has both curated solutions for the common use cases, and the extensibility to make custom visualizations/transforms for more specific needs. Many of the Stack's capabilities are free and open, meaning you can see the code you're going to run, and set up and operate most of these on your own. If you want to get up and running quickly, Elastic also has a cloud offering that can have you testing things out in minutes.
A brief overview of some of the Observability capabilities:
- Synthetics - powerful endpoint monitoring. You can perform lightweight
http
/tcp
/icmp
checks on both internal and external endpoints. Additionally, users can leverage powerful full-browser checks for things like logins and other webpage functionality checks. These checks also gather detailed performance metrics that can highlight real-world deficiencies seen by end users, along with screenshots of each check for better troubleshooting. - APM - a full-fledged application performance monitoring solution that provides visibility into your code performance on the server.
- RUM - real user monitoring, helps track the experience of different users/devices while they're on web pages
- Logging - centralize and analyze your infra's log data in a single pane of glass.
- Metrics - monitor the performance of underlying infra like servers, docker containers, K8s pods, etc.
Elastic also has offerings built around security monitoring and incident tracking, machine learning that makes it easier to catch deviations from normal performance (think traffic spikes), and more.
If you have specific questions about your use case feel free to post back here and I'll do my best to answer. Thanks again for your interest in the Elastic Stack!
Thanks for detailed response
Is APM allow me collect application metrics or provides some form of integration with existing Prometheus instances and/or integrations?
I have difficulties with understand when I should use beats, and when agents. Is agents will replace beats in future?
Hi @musialny - so, there are individual Beats like Heartbeat, Metricbeat, Filebeat, etc., each has its own purpose. Elastic Agent is, in essence, a bundling of all the Beats functionality for multipurpose use, and it's specifically designed to work with Fleet, our centralized management system.
Whether you use individual Beats or a Fleet/Agent deployment depends greatly on your use case and how many things you're planning to monitor. If you're just getting started, you can play around with whichever Beats you want to try out, or you can set up a cloud trial and get up and running with fleet and agent in a few minutes. The advantage of cloud is you won't need to set up or configure Kibana or Elasticsearch, and getting Agent running and enrolled is as simple as pasting a docker command into a terminal.
APM allow me collect application metrics or provides some form of integration with existing Prometheus instances
I would ask the Prometheus question to the APM folks. Elastic is no stranger to Prometheus but I won't be able to provide you as good of an explanation as they can.
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.