ICYMI – Network Monitoring and InfluxDB

Navigate to:

At this year’s InfluxDays event, the capabilities of InfluxDB took center stage. It’s not enough to simply deploy a technology platform and hope people will use it. This isn’t a Kevin Costner movie. That’s why it’s helpful to talk about specific use cases, their typical challenges, and how InfluxDB can address those challenges. Fortunately, that’s just what Influxer Charles Mahler did for network monitoring.

Here are some of the highlights from Charles’ presentation.

It all starts with data collection

In order to monitor anything, you need to have data about that thing. Obtaining good, clean data is, therefore, one of the most critical aspects of network monitoring. It’s also one of the biggest challenges. This is especially true at the enterprise level, but any situation that involves legacy hardware, different operating systems, and/or a variety of sources increases the challenge. Many hardware vendors require users to accept a proprietary format or protocol, which can create more work when collecting data.

Oftentimes, the raw data collected doesn’t fit the bill of ‘good’ or ‘clean,’ which means that users either must deal with dirty data or create new processes to standardize that data across their system. If you want to enrich raw data with data from another source, that can create additional challenges.

Data storage

Of course, once you collect data, you need to keep it somewhere. Here you need to balance cost with complexity. Some use cases require high granularity data, but this comes with added storage costs. It’s important to find the right metric intervals so that you can get the insights you need and afford the storage for your data, too. Data storage becomes complex at scale, so this is another factor to consider when thinking about your network monitoring data. More devices mean more data and more storage requirements.

Analyzing your data

Only after collecting and storing your data can you do any kind of analysis of it. You need to make sure that you have enough compute to handle the queries you want to run on your size data set. Otherwise, those queries will be slow, which can cause more significant issues downstream. At the same time, there’s a need to ensure that data analysis isn’t just a black box that a limited number of people understand. Providing tools that enable a wide range of users to engage with your data and to gain value from it becomes really critical.

InfluxDB to the rescue

Knowing the challenges that people face when building network monitoring solutions helps to understand how InfluxDB addresses these challenges and provides the type of observability and access to data that users want.

Data collection made easy

Collecting data is a point of emphasis for the InfluxDB platform. Telegraf is our open-source data collection agent. It has hundreds of plugins so you can collect data from virtually any source. (We even have an InfluxDB University course that covers how to write your own Telegraf plugin!) You can use plugins to amend, transform, and clean data before it even hits your datastore. So, if you have thousands of devices you need to track, Telegraf has you covered.

In addition to Telegraf, InfluxDB has client libraries in over a dozen languages, so you can write code in a language you’re comfortable with, accelerating the development process.

Data storage that scales

InfluxDB Cloud is our fully managed SaaS solution and it automatically scales to handle your storage needs. Whether you need to retain high granularity data, coarser data, or a combination of the two, InfluxDB can store it.

Data analysis tools you can use

When it comes to query performance, InfluxDB is purpose-built for time series data and its unique workloads. Users that come to InfluxDB from other database types see a huge jump in query performance. As insights come faster, you can iterate your network monitoring applications and dashboards more quickly.

Speaking of dashboards and visualizations, InfluxDB lets you build those quickly. And these are tools that anyone can use and gain value from, not just subject matter experts. InfluxDB has a visual query builder so anyone can build dashboards and uncover new insights. This allows teams to expand the scope of observability and innovation.

A big addition to the InfluxDB platform, as part of its new time series engine, is support for SQL (specifically the PostgreSQL wire protocol). Now seasoned SQL users can leverage their know-how to query data in InfluxDB. SQL support also makes InfluxDB more extendable, as it can communicate with other PostgreSQL-based solutions, and further democratizes the development and data analysis processes.

You don’t just have to take our word for it. Companies like Cisco and Red Hat (and many more) use InfluxDB to power network monitoring solutions.

This recap just scratches the surface regarding network monitoring and InfluxDB. For more details about the challenges of network monitoring and how InfluxDB solves them, check out the full presentation.