© Atos Spain S.A. 2016
The Monitoring Infrastructure is a component of the European Project TANGO (http://tango-project.eu ).
The responsibility of this component is twofold. On one side it must provide running applications with metrics of the status of different devices, such as CPU, GPUs, FPGAs, etc. This information should be both energy consumption and performance status. On the other side it should also provide historical statistics for device metrics for latter analysis of upper components of the TANGO architecture.
This component relies on Collectd, which will be used together with some custom plugins done specifically for TANGO, in order to get and store energy consumption and performance values from the applications running in the different environments.
Available collection of probes to measure energy in heterogeneous hardware for the TANGO project:
- memory used
- processes running
- memory used
The monitoring Infrastructure is composed by the following components:
Collectd & plugins
Collectd is a daemon which collects system and application performance metrics periodically and provides mechanisms to store the values in a variety of ways. The TANGO Collectd plugins provide metrics from specific devices like the NVIDIA GPUs and the XEON PHI processors. The metrics gathered from this component are stored in the InfluxDB instance.
InfluxDB is used as a data store for the metrics gathered from Collectd. The REST API and the Graphana applications connect to this tool in order to present the processed information.
Graphana is a data visualization and monitoring tool with support for Graphite, InfluxDB and many more databases. This component is used as the GUI of the monitoring component.
The main purpose of the REST API application is to offer to other TANGO components the information they need from the metrics gathered from Collectd.
View REST API specification to see the available methods.
Installation Guide (ubuntu, 64-bit systems)
- ubuntu 64-bit system
- Java 8
- Leiningen 2.0.0 or above installed.
2. Install Collectd
sudo apt-get update sudo apt-get install collectd collectd-utils
2.1. Install plugins
View Collectd - Readme file to see how to compile and install the plugins.
3. Install InfluxDB
wget https://dl.influxdata.com/influxdb/releases/influxdb_0.13.0_amd64.deb sudo dpkg -i influxdb_0.13.0_amd64.deb sudo service influxdb start sudo service influxdb statu
4. Install Grafana
wget https://grafanarel.s3.amazonaws.com/builds/grafana_3.0.4-1464167696_amd64.deb sudo apt-get install -y adduser libfontconfig sudo dpkg -i grafana_3.0.4-1464167696_amd64.deb
5. Collectd & InfluxDB & Grafana setup
See Monitoring with Collectd, InfluxDB & Grafana for more information about how to setup the three components.
6. Install and configure the REST API component
View rest-api - Readme file to see how to install and configure the REST API component.
Relation to other TANGO components
The other TANGO components connect to the monitoring Infrastructure through the REST API application:
heterogeneous parallel device clusters
The Monitoring Infrastructure project is licensed under a GNU General Public License, version 3.
rest-api component depends on a set of libraries and plugins licensed under the Eclipse Public License v1 and the Eclipse Public License v2. These libraries are used and imported through leiningen or maven, and they don't belong to the distribution of this project. This means that the original source code of these libraries was not modified. These libraries are the following: