Yesterday, Chronosphere, the cloud native observability platform, raised an additional $115m Series C at a $1.6bn valuation. Last quarter, the company tripled ARR, with over 145% NRR, and is one of the fastest growing SaaS companies globally.
This could be a problem for Datadog.
Chronosphere is built by Former Uber engineers who experienced the chaos of managing data as Uber moved to a cloud native stack.
When Uber moved to cloud native, they saw exponential data growth on the metric side. The reason is, say you're running 1,000 VMs, and now you've moved to cloud native. You’ve been able to consolidate, so now you're running 800 VM. Now you're running 10,000 containers on top of those VMs, and you're collecting metrics from 10,000 different individual resources, whereas before, you were only collecting metrics from 1,000. When they made that shift, they were hitting above a billion metrics a second, as an application, and they needed a way to have something scalable.
Chronosophere’s pitch is to reduce the cloud monitoring bill of cloud-native companies:
As an example, our average customer can reduce metrics that are incoming by over half, and we bill based on persisted metrics, which is very different than the rest of the industry. The rest of the industry bills are based on incoming metrics. If you want to control your bill, you have to set up a pipeline in front of your metric system that weeds out data but causes lags. Many companies are going in this direction, but they've all been struggling to go in this direction. They're doing that because their vendors, New Relic, Datadog, and Dynatrace, all bill on an incoming metric system.
This interview with a Chronosphere sales team member explores how the observability landscape is changing, the role of Prometheus, and risks to Datadog.