# Reduce Prometheus memory usage when scraping Node metrics

**URL:** <https://radixtalk.com/t/reduce-prometheus-memory-usage-when-scraping-node-metrics/66>\
**Category:** Validators\
**Tags:** howto, prometheus\
**Created:** [25 December 2021 14:23 UTC](https://radixtalk.com/t/reduce-prometheus-memory-usage-when-scraping-node-metrics/66 "2021-12-25T14:23:45Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![Stuart](https://radixtalk.com/letter_avatar_proxy/v4/letter/s/cab0a1/32.png) [@Stuart](https://radixtalk.com/u/Stuart)\
**Post date:** [25 December 2021 14:23 UTC](https://radixtalk.com/t/reduce-prometheus-memory-usage-when-scraping-node-metrics/66/1 "2021-12-25T14:23:45Z")

</div>

I had a look at my Prometheus time series database stats and noticed that the `version_accumulator` label from the `nodeinfo` metric has the highest cardinality by a significant margin. This is because the value changes every single metrics scrape. It also has the highest memory usage compared to the second highest label (x15 more). You can read about why high cardinality labels are bad at this link:

[https://www.robustperception.io/cardinality-is-key](https://www.robustperception.io/cardinality-is-key)

Here’s the snippet I added to radix metrics job in `prometheus.yml` to drop the `version_accumulator` from being added to the time series database:

```auto
    metric_relabel_configs:
      - regex: version_accumulator
        action: labeldrop

```