How to run InfluxDB on Google Cloud

With InfluxDB Cloud for Google Cloud, GCP customers can use our leading time series data platform on Google infrastructure. This lets you address a wide range of use cases: observability, server monitoring, IoT sensor data tracking, real-time customer analytics, application performance metrics, network monitoring, security threat detection, and financial market analysis. Whatever data you might want to examine, monitor, and act on over time, InfluxDB Cloud provides you with a reliable foundation that requires minimal effort to set up.

What is InfluxDB Cloud for Google Cloud?

InfluxDB for Google Cloud 2.0 is a serverless cloud platform that’s purpose-built for time series data. That’s a lot of buzzwords, but we’re trying to pack in a number of points. Let’s break them down.

By serverless, I don’t mean that InfluxDB Cloud doesn’t run on servers. (Obviously it does!) Instead, I mean that you don’t need to worry about sizing your database servers to match your workloads. You don’t have to worry if you bought too much capacity or too little, or about having to migrate off of one server size to the other. InfluxDB Cloud automatically handles the elastic scaling required at the backend, so you don’t have to. This way, migration from one size database server to another is never required. Our pricing isn’t based on the size of the machines you choose; it’s based on the size of your data.

By cloud, I mean that enhancements to InfluxDB Cloud 2.0 come within days, if not hours, of when they’ve been committed in our open source database, InfluxDB. This lets you take advantage of new innovations as soon as they’ve been tested and publicly deployed by our team. It also frees up time from your sysadmins and SREs to focus on other parts of your data infrastructure.

By platform, I mean that InfluxDB for Google Cloud today has all the core capabilities of InfluxDB 2.0 database: time-series data storage and querying, processing in the background, numerous integrations with third-party services (including those from Google Cloud), collection agent configuration, and highly configurable dashboards and alert processing.

By purpose-built, we mean a database that handles the relentless scale of time-stamped data generated by modern microservices, devices, and sensors — something that general-purpose databases can’t do. Think of other types of use cases, such as order entry or CRM. For any single customer, writes might come once every few minutes, when someone enters an order or updates a customer record. In contrast, for time series data, you might have dozens of writes every second coming from different monitoring agents or sensors. The shape of the data is simpler than relational data, but it’s a firehose of information.

How does InfluxDB Cloud integrate with Google Cloud?

InfluxDB Cloud for Google Cloud runs in Google Cloud’s central region, us-central1, in Iowa. InfluxDB Cloud now integrates with a wide range of Google services, bringing observability to your entire Google Cloud stack:

• Stackdriver: Telegraf, the data collector for InfluxDB, can use its Stackdriver Input Plugin to grab metrics from 40 different Google Cloud services and store them in InfluxDB for dashboarding with Chronograf and analysis with Flux, the new data scripting and query language for InfluxDB. There’s also a Telegraf plugin to output data to Stackdriver.
• Google Kubernetes Engine: You can use Telegraf for GKE monitoring, thanks to Telegraf’s Kubernetes plugin, which lets you talk to the Kubelet API and gathers metrics about the running pods and containers for a single host. You can also use the Telegraf Kube Inventory plugin, which generates metrics derived from the state of Kubernetes resources such as daemonsets, deployments, nodes, and more. And there’s the Telegraf Prometheus plugin, which gathers metrics from HTTP servers exposing metrics in Prometheus format.
• PubSub: Telegraf can ingest messages from PubSub into InfluxDB, and send data out of InfluxDB to Google Cloud PubSub using its Cloud PubSub output plugin. From there, you can send data from PubSub to BigQuery, and from there to Google AI platform for further analysis.
• Cloud Build: Telegraf can monitor containers used by Google Cloud Build, thanks to its Docker plugin and Docker Log plugin.
• Compute Engine: Telegraf can monitor virtual machines spun up by Google Compute Engine, using its monitoring plugins to capture metrics for server CPU, disk, memory, network, processes, swap, and system uptime.
• IoT Core: Telegraf integrates with Google IoT Core, to help you collect metrics from devices and sensors.
• MQTT: Your devices can use an MQTT bridge to communicate with Cloud IoT Core. Telegraf can consume MQTT topics, using its MQTT input plugin. Telegraf can also output data to MQTT as well.
• BigTable: Flux can import data from BigTable, and do joins between BigTable, InfluxDB, and other data sources. (If you’re not familiar, Flux is our new open-source data scripting language, designed for working with time series data.)

Notice that InfluxDB and Telegraf can be used not only to ingest data from Google Cloud, but to send data to Google Cloud as well, via the Telegraf output plugins for PubSub, MQTT, and Stackdriver. This is an important thing to look for in monitoring solutions — you don’t want your monitoring data to get trapped in a “roach motel” that limits your options.

Why use InfluxDB Cloud on Google Cloud?

Real-time observability: Telegraf can collect metrics as frequently as once per second. InfluxDB dashboards update every five seconds. InfluxDB checks run every five seconds. InfluxDB also has deadman checks, which fire when a group of agents stop reporting. All these capabilities mean that once something goes wrong, your SREs will know about it sooner, which makes it easier to achieve your service level objectives (SLOs).

Flexible monitoring: Every dashboard chart in InfluxDB, and every InfluxDB check, has a Flux script behind it and can have multiple queries, joins across datasources (including Stackdriver, BigTable, and PubSub), variables, comparisons, regular expressions, and statistical functions. This gives you practically unlimited flexibility to fine-tune your monitoring to meet your needs.

Time series AI: Google Cloud AI Platform has some incredible AI and machine learning services, such as TensorFlow, TPUs, and TFX, which are ideal for analyzing the vast amounts of time series data produced by sensors, devices, and app infrastructure. Once you send your time series data from InfluxDB to BigTable, BigQuery, or PubSub, it can feed Google AI platform to generate insights.

How to sign up for InfluxDB Cloud on Google Cloud?

You can sign up for InfluxDB on Google Cloud through InfluxData’s website. When prompted for your cloud provider, just pick Google Cloud Platform as shown below.

What’s next?

If you’re already using InfluxDB Cloud — enjoy these new changes. If you’re not — sign up! Then be sure to check our quick start guide and ask the InfluxDB community and Slack for help.