'Filter values before aggregating to a number on DataDog "query_value" visualization

I have a counting metric in DataDog, that has positive and negative counts. When using a "timeseries" visualization to show only the positive counts, I simply just apply a clamp_min(query1, 0), and it works.

Essentially, the JSON looks like below for the "timeseries"

{
    "viz": "timeseries",
    "requests": [
        {
            "style": {
                "palette": "dog_classic",
                "type": "solid",
                "width": "normal"
            },
            "type": "line",
            "formulas": [
                {
                    "formula": "clamp_min(query1, 0)"
                }
            ],
            "response_format": "timeseries",
            "queries": [
                {
                    "query": "sum:my_counting_metric.as_count()",
                    "data_source": "metrics",
                    "name": "query1"
                }
            ]
        }
    ]
}

When I try to do this same thing, and aggregate all these positive numbers that I've filtered for on a "query_value" viz, I can't find a single way to do anything similar because it sums all the counts to a number before making it possible to apply filters like clamp_min above. Ideally, I would like to do something like below if possible.

{
    "viz": "query_value",
    "requests": [
        {
            "response_format": "scalar",
            "formulas": [
                {
                    "formula": "query1"
                }
            ],
            "queries": [
                {
                    "query": "sum:clamp_min(my_counting_metric.as_count(), 0)",
                    "data_source": "metrics",
                    "name": "query1",
                    "aggregator": "avg"
                }
            ]
        }
    ],
    "precision": 2,
    "autoscale": true
}

Is there any way to do this? I'm feeling pretty stuck and would greatly appreciate any help! 🙏🏻

datadog


Solution 1:[1]

I was able to get in contact with DataDog on this question.

They basically told me this behavior mentioned in my post is expected because the "query_value" viz runs a procedure called point reduction. This is why the clamp_min function had no effect. To put it short, there's not a way to filter/map before the execution of the initial aggregating function on a "query_value" viz.

Sources

This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.

Source: Stack Overflow

Solution Source
Solution 1 DougCal

Related Questions