Hey everyone, I'm trying to build a business case for my team on observability tools, and I've hit a wall trying to compare the actual ingestion costs for Honeycomb and Datadog.
Our current setup is pretty basic - we're sending logs and some custom metrics from Docker containers running on a few VMs. I'm looking at maybe 150GB of observability data per month to start. Both platforms talk about "ingested data," but the pricing models seem totally different and I'm having a hard time mapping them to a simple "cost per GB."
For Datadog, I see they charge per "million log events" and per "custom metric." How do I even estimate how many "events" are in a gigabyte of JSON logs? 😅 And with Honeycomb, it's all about "events" and "MTU" (Million Trace Units). Their calculator helps, but I'm not sure if I'm modeling our data correctly.
Could someone walk me through a real-world example? Like, if I have a simple Flask app logging HTTP requests (with fields for `status`, `duration`, `path`), what would the cost breakdown look like for 1GB of that data on each platform?
I'm also curious about how things like sampling or dropping fields *before* it gets sent affects the bill. Is that a common strategy to control costs? We're just starting out, so I want to avoid any nasty surprises in a few months.
Learning by breaking