Items Prometheus-Grafana Zabbix
Zabbix has a core written in C and a web UI
based on PHP. Also it uses "agents" (client-
Technology implementation Written in GO language
side programs) written in C (Implemented
in C for better lower memory footprint)
The default MySQL Database is used for
Performance, monitoring data
It stored the data locally internal database storage, it also supports MySQL,
collection and storage
PostgreSQL, Oracle, or SQLite
Yes. Prometheus has native support for
Support for Docker and Kubernetes Yes
containers and Kubernetes.
Prometheus and Grafana have a variety of
Zabbix has very limited support for 3rd
Plugins and Integration opensource plugins to support various
party integration
infrastructure monitoring
For Zabbix, you have to use SQL(Structured
Prometheus has PROMQL for querying the
Querying the data for analysis Query Language) to fetch the metrics from
metrics
either Mysql or Postgres
Grafana has really good support for custom
dashboard also there are various
opensource ready use dashboard which can
be imported. Prometheus offers a basic tool Zabbix also has a really good Dashboard
for exploring gathered data and visualizing but does not support importing dashboard
Dashboard and Visualization
it in simple graphs on its native server and from other developers. Zabbix offers its
also offers a minimal dashboard builder. But own web UI for visualization.
Prometheus is and is designed to be
supported by modern visualizing tools like
Grafana
Prometheus and Grafana has really good Zabbix also has a very good support for
Alerting support for alerting SMS, Microsoft alerting on SMS, Microsoft Teams, Slack,
Teams, Slack, Email Email
Prometheus uses HTTP with Protocol Zabbix uses its own TCP-based
Communication Buffers (+ text format for ease of use with communication protocol between agents
curl). and a server.
Zabbix by default uses a "pull" model when
a server connects to agents on each
Prometheus uses a "pull" model when a
monitoring machine, and agents
server gathers information from client
periodically gather the information and
Data Model machines. But Prometheus Push Gateway
send it to a server. The alternative is
may be used in cases when a "push" model
"active checks" mode when agents
is needed.
establish a connection with a server and
send data to it when it need.
An open-source service monitoring system
Track, record, alert and visualize
and time series database, developed by
performance and availability of IT
SoundCloud. Prometheus is a systems and
resources. Zabbix is a mature and effortless
service monitoring system. It collects
Usage enterprise-class open source monitoring
metrics from configured targets at given
solution for network monitoring and
intervals, evaluates rule expressions,
application monitoring of millions of
displays the results, and can trigger alerts if
metrics.
some condition is observed to be true
use the PROMQL to fetch the cpu usage by https://www.zabbix.com/upgrade_services
Pricing putting the
container_cpu_usage_seconds_total starting from $ 1500 (USD)
Items Prometheus-Grafana Zabbix
parameter in the PromQL. You can also
create a New Dashboard where you can
measure the CPU usage in an hourly
manner
(sum(container_cpu_usage_seconds_total)
/ 60 / 60 ) * 0.0445
Whatever value is return from the above
expression can be multiplied by the hourly
price of the cloud environment which will
give you a rough estimate of the cost
involved in running Prometheus and
grafana.
Complex. It does not come with features
such as alerting or complex data
Simple. Zabbix has more pre-installed
visualization out of the box, so
Installation features, making its installation and setup
organizations must separately install and
faster.
integrate these features and manually
configure files.
Prometheus tends to perform very well, Zabbix performs very well up to around
even on exceptionally large networks. This 10,000 nodes. After 10,000 nodes, Zabbix
Scalability is due to its native time-series database, starts to suffer under the weight of its
cluster management and overall simplicity external relational database and its lack of
of its architecture. reliable management for larger clusters.
