Ajay Tripathy

Hey @nicolodallavalentinaclearscore , I think this is something we'd take a PR for if interested.

We actually just got a contribution that potentially might help! https://github.com/opencost/opencost/pull/1792/files

Maybe our docs need clarification. https://www.opencost.io/docs/prometheus says `Average number of CPUs requested/used over last 1m` Whereas it should say something like `Max of the (CPU Request,CPU Usage) over the last...

Thanks for that feedback. The decision was made to use a single metric that could be multiplied by price to get total costs for ease of use. You can price...

No, we rely on these metrics in kubecost/opencost to compute allocation cost. And no, opencost does not, though of course you can whitelist what you want on the prometheus side.

Hey @dudicoco , that's a pretty sizable refactor and not currently on the roadmap, but thanks for the feedback.

Here's an example API. ``` GET /model/allocation?window=7d...&includeProportionalAssetCost=node { "cluster/node/namespace/pod/container": { ...all existing fields, "proportionalCost": { "node1": { "cpuCost": 2.07, "cpuPercentage": 0.004321, "ramCost": 0.22, "ramPercentage": 0.083745 }, "node2": { "cpuCost": 2.07,...

@wamak9 we'll look into adding this into the UI-- I was just wondering if a visual display of the data we're showing here is more or less what you're looking...

Hey @dudicoco the kubecost API has a number of caching layers and optimizations for long running queries. What's the size of your cluster?

@dudicoco one solution would be to run the shorter queries of 1d length or lower on opencost and export these as a batch job to an external data source. That's...