aws-samples / aws-samples/aws-batch-operational-dashboards

Clarification on Split Cost Calculation

Open
#36 21 comments 0 reactions 1 assignee Claimed by @mhuguesaws View on GitHub
bug
Dominant language
Python
Stars
15
Forks
2
PR merge metrics
No merged PRs in 30d

Description

Quick question about how the Batch costs are calculated in the dashboards. I noticed a bit of a mismatch and wanted to check if I'm missing something obvious.

So from what I understand, the split cost for Batch is calculated based on the Compute and memory (used + unused) of the EC2 instances. But when I compare the numbers across tools, they don’t quite line up. ([AWS Blog Source](https://aws.amazon.com/blogs/aws-cloud-financial-management/la-improve-cost-visibility-of-containerized-applications-with-aws-split-cost-allocation-data-for-ecs-and-batch-jobs/))

For example, when I check Cost Explorer for EC2-Instance usage in February, I see:
- Cost Explorer EC2-Instance total: $1,580.66

But in Grafana, the sum of team costs for the same period comes out to:
- Grafana total: ~$4,627

I also queried Athena directly to double-check, and it aligns more with what Grafana is showing — so not sure what Cost Explorer is excluding (or what we're maybe double-counting?).

Just wanted to clarify:

- What services are included under Batch in the split cost allocation?
- If it’s really just CPU and memory (used + unused) from EC2, do you have any idea why this difference might show up?

Appreciate any insight — thanks!

![Image](https://github.com/user-attachments/assets/84687f01-18ac-44f8-96b2-89ad8618b3d7)

![Image](https://github.com/user-attachments/assets/2c9c544a-f2ef-4ebe-a152-15770153da84)

Contributor guide

Open the contributing guide

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.