Prometheus & Grafana Requirements — Per Target Growth Model

Version 1.6.5 | Updated: March 10, 2026

Prometheus & Grafana Requirements — Per Target Growth Model

This document describes how to calculate the infrastructure requirements for Prometheus 3.5.0 and Grafana v12.1.0 when adding new servers and exporters. It provides a baseline setup, a per-target growth impact matrix, and scaling thresholds.

1. Current Baseline

  • Prometheus

    • 4 vCPU

    • 16 GB RAM

    • NVMe SSD ≥ 500 GB (≥5k IOPS)

    • Retention: 90 days

    • Scrape interval: 60s

    • Current load: ~1.74M active series, ~200 GB TSDB

  • Grafana

    • 2 vCPU

    • 4 GB RAM

This setup is stable for the current workload.

2. Per-Target Growth Impact

The following table shows the average number of series and the expected storage impact per target (90 days retention, 60s scrape interval).

Exporter / Job Avg Series per Target Storage @90d (≈) Notes

JmxExporter (HC)

~847k

~194 GB

🔴 Main driver of load

JmxExporter (ACS)

~27

~0.05 GB

Negligible

MySQL Exporter

~16.3k

~32 GB

Heavy

Process Exporter (ACS)

~7.2k

~14 GB

Medium-heavy

Process Exporter (HC/DB)

0.4—​0.6k

<1 GB

Light

Node Exporter

0.7—​1.3k

~2 GB

Stable

Northbound API

1.1—​2.4k

2—​4.5 GB

Variable

Service API

0.5—​0.8k

~1 GB

Light

Provision API

0.4—​0.6k

~1 GB

Light

Subscription API

~0.3k

~0.6 GB

Very light

Wildfly Exporter

0.2k

~0.4 GB

Scrape latency up to 5s

FT System Metrics

0.8—​0.9k

~1.6 GB

Light

Prometheus self-scrape

~1.3k

~2.6 GB

Always present

3. Growth Calculation Rule

  1. List the exporters that will run on the new server.

    • Example (HC server): JmxExporter (HC) + node-exporter + process-exporter

    • Example (ACS server): JmxExporter (ACS) + acs-metrics + ft-system-metrics + node-exporter + process-exporter + northbound-api + provision-api + service-api + subscription-api + wildfly-exporter

  2. Take the values from the table above.

  3. Sum the series and storage impact.

  4. Compare with thresholds to decide if scaling is needed.

4. Scaling Thresholds

  • <2M active series → Baseline (4 vCPU / 16 GB) is sufficient.

  • 2—​4M active series → Scale Prometheus → 8 vCPU / 32 GB RAM, SSD 1 TB.

  • 4—​8M active series → Scale Prometheus → 16 vCPU / 64—​96 GB RAM, SSD 2—​4 TB.

  • >8M active series → Scale Prometheus → 32 vCPU / 96-124 GB RAM, SSD 4-8 TB.

Grafana load is mainly driven by the number of users/dashboards rather than series count.

5. Example: Adding DevOps-HC3 + DevOps-ACS3

DevOps-HC3

  • JmxExporter (HC): ~847k series, ~194 GB

  • node-exporter: ~1k series, ~2 GB

  • process-exporter (HC): ~0.4k series, ~0.8 GB

  • Total (HC3): ~848k series, ~197 GB

DevOps-ACS3

  • JmxExporter (ACS): ~27 series, ~0.05 GB

  • acs-metrics: ~180 series, ~0.35 GB

  • ft-system-metrics: ~900 series, ~1.6 GB

  • node-exporter: ~1.3k series, ~2 GB

  • process-exporter (ACS): ~7.2k series, ~14 GB

  • northbound-api: ~2k series, ~3 GB

  • provision-api: ~500 series, ~1 GB

  • service-api: ~600 series, ~1 GB

  • subscription-api: ~315 series, ~0.6 GB

  • wildfly-exporter: ~220 series, ~0.4 GB

  • Total (ACS3): ~11k series, ~24 GB

Combined Growth

  • Grand Total (HC3 + ACS3): ~860k series, ~220 GB

Impact

  • Current: ~1.74M series, ~200 GB

  • After adding HC3 + ACS3: ~2.6M series, ~420 GB

Impact:

  • This represents a ~50% increase in active series

  • Action: Scale Prometheus to 8 vCPU, 32 GB RAM, SSD 1 TB

  • Grafana can remain as-is