InfluxDB Cloud is now beta on Microsoft Azure and we’d like you to try it out so we can work to make sure it’s ready for general availability (GA). If you’re already familiar with InfluxDB, head over to cloud2.influxdata.com to sign up for your free account. And if you’re not, read on for how to get started.
With InfluxDB Cloud on Azure, you can:
- Manage time series data from Azure-connected IoT devices and sensors
- Monitor applications and infrastructure running on Azure
- Connect InfluxDB to Azure services such as Azure Application Insights and Azure Monitor
As an auto-scaling database cluster, InfluxDB Cloud has nearly infinite scalability and requires you to do almost no database management of your own.
Our new Azure service extends our existing availability in AWS Oregon and Frankfurt and Google Iowa. (That’s as of when this post was written. Here is the current list of InfluxDB Cloud Regions.) Because much of the interest for InfluxDB Cloud on Azure came from our European users, we decided to start this service in the Azure Western Europe datacenter in Amsterdam.
Wait… this sounds familiar
If you’re getting a case of deja vu, it’s because a couple of weeks ago, we announced the GA of InfluxDB Enterprise on Azure. So, how is this different?
InfluxDB Enterprise on Azure — announced two weeks ago — is based on open-source InfluxDB 1.8, and adds role-based access control (RBAC), clustering, and identity and access management. It runs in 33 Azure locations worldwide, has integrated billing through Azure Marketplace, and is ready for production workloads.
InfluxDB Cloud on Azure beta — today’s announcement — is based on open-source InfluxDB 2.0, and is a serverless time series data platform. By serverless, we mean that you don’t need to worry about picking the size of your InfluxDB server in terms of memory, disk, or compute.
Instead, 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, and you never have to worry about buying too much or too little capacity. Just pay for what you use — no more, no less.
What is the TCO of InfluxDB Cloud?
InfluxDB Cloud on Azure comes with instant-on, automatic backup, automatic load balancing, automatic upgrades, security hardening and is managed 24x7x365 by our site reliability engineering team. All the complexity of running a time series database is done for you.
How much time will this save you? That depends on the size of your deployment. That said, one customer told me recently that, for them, InfluxDB Cloud frees up about half an FTE (full-time equivalent) in terms of system administration effort. Looking up the average base salaries for site reliability engineers or system admins in the UK, Germany, and San Francisco, and assuming 50% additional for bonus, taxes, benefits, office space and equipment, we see that saving half an SRE is a significant amount:
|Location||Average salary||SRE full cost (est.)||Half SRE FTE (est.)|
|UK||GBP 70,000||GBP 105,000||GBP 52,500|
|Germany||EUR 69,000||EUR 104,000||EUR 52,000|
|San Francisco||USD 105,000||USD 158,000||USD 79,000|
Your mileage will vary of course, but this should give you a sense of what you’d save by moving to a fully-managed database service like InfluxDB Cloud.
Is InfluxDB Cloud on Azure production-ready?
Not yet! Because InfluxDB Cloud on Azure is beta, we don’t yet recommend it for production workloads, and because of that, only the free tier is available for now. The point of this release is to have folks like yourself try it out with development and trial workloads. This will help us ensure the service is rock-solid before we release it as production-ready GA and offer a paid tier.
We don’t yet have a date for the GA of InfluxDB Cloud on Azure, but we’ll announce on this blog when it’s ready.
InfluxDB Enterprise vs. InfluxDB Cloud
Here’s a summary of the differences between InfluxDB Enterprise and Cloud:
|InfluxDB Enterprise on Azure||InfluxDB Cloud on Azure beta|
|API compliance||InfluxDB OSS 1.x||InfluxDB 2|
|Azure regions||33 regions worldwide||1 region – Amsterdam|
|Azure marketplace billing||Yes||No|
Why only one Azure region?
All of the great features of InfluxDB Cloud on Azure — serverless, elastic scaling, instant-on, etc. — require a constantly running cluster of InfluxDB open source, Kubernetes, Docker, and other technologies. As you know if you run your own cloud infrastructure, these are hardly free! This level of cloud infrastructure investment isn’t unique to InfluxDB — they are necessary for any high-availability, elastically scaling cloud service. So, we are rolling out cloud regions where user demand is highest. This focus on efficiency is reflected in our prices for InfluxDB Cloud, which charges fractions of a penny per gigabyte of storage-hour.
We welcome you to share your ideas on which regions we should expand to next, whether on Azure or another cloud provider, in this survey.
How to reduce Azure data transfer
Given that InfluxDB Cloud runs in only one Azure region, you might be wondering about data transfer if your application is outside of their West Europe region. But that’s unlikely to be an issue for a few reasons:
For Azure West Europe, inbound data transfer is free. So, no issues there.
But even if inbound data transfer is free, you might still want to reduce the volume of inbound data to reduce latency. Here are a couple of ways you can do that:
- Reduce your sample frequency. In Telegraf, this is as simple as setting one line in your telegraf.conf file. For example,
interval = "60s"will set your sample frequency to once every 60 seconds;
interval = "5m"will set it at once every 5 minutes.
- Turn on compression. In your telegraf.conf file, add
content_encoding = "gzip". This will speed up data transfer by about 5x.
You get 5 GB of outbound data transfer per month from the Azure West Europe region, which works out to 2 KB/second continuously, 24 hours a day, 30 days a month. That’s a sizable amount of data for most applications, especially if you manage your query results size using the Flux aggregateWindow() function. For instance, this line of Flux will take each 15-minute segment of your data, calculate the mean value, and return the result:
|> aggregateWindow(every: 15m, fn: mean)
aggregateWindow() will take dozens, hundreds, or even thousands of rows of time series data and reduce them to a single row, greatly reducing your outbound data.
How to get InfluxDB Cloud on Azure
Go to cloud2.influxdata.com to get started. You’ll see a screen like this:
Later in the registration process, you’ll be prompted to pick where to host your data. Once you pick Azure, you’ll instantly get access to InfluxDB Cloud. There’s no database-on-a-VM that you need to wait to start up. Because InfluxDB Cloud is a multi-tenant service, you can immediately get to work. That’s what we mean by instant-on.
Now that you’ve signed up for InfluxDB Cloud on Azure, here are some new tips and content on how to get started with it.
How to get data into InfluxDB Cloud
Once you’ve completed the signup process above, you’ll want to add some data to your InfluxDB Cloud instance. To do that, check out this video:
After you’ve watched the video above, check out our Getting Started Guide for InfluxDB Cloud.
How to find your InfluxDB Cloud URL
The video demo above uses InfluxDB Cloud on AWS, but the instructions are the same for InfluxDB on Azure beta. Just use the Azure URL on this page.
Speaking of URLs, here’s another trick: click the gear icon in the upper right corner of our InfluxDB Cloud 2 documentation, so you see this screen:
Check off “Azure West Europe,” as shown above. That way, all URLs across our InfluxDB Cloud 2 doc will automatically be customized to point to Azure. Pretty slick work by our docs team!
How to share InfluxDB Cloud
Check out InfluxDB Cloud multi-user support to see how you can share InfluxDB Cloud with your teammates — it works for all InfluxDB Cloud regions, including Azure beta. “Owner” is currently the only permission level for multi-user support, but as you can see below, we’ve designed the UI to accommodate additional user levels as they are delivered by our engineering team.
To learn more about multi-user support, check out this video:
Your feedback, please
We are releasing the InfluxDB Cloud on Azure beta to ensure that it’s rock-solid once we go GA. We’d love it if you kicked the tires, run it through paces, and shared your feedback by emailing the InfluxDB Cloud product team at [email protected].