CompareUpdated 2026-09-22
Infrastructure monitoring: SolarWinds vs SigNoz vs AMP with Grafana vs Datadog Infrastructure
SolarWinds and Datadog Infrastructure charge for every node or host they monitor, whether or not anyone reads the data, while SigNoz and Amazon Managed Service for Prometheus charge on the data itself and can keep it in infrastructure or a cloud account you own.
24cells, every one sourced and dated
Side by side
From the vendors' own pages
- SolarWinds Observability SaaS pricing
- SigNoz pricing
- Amazon Managed Service for Prometheus pricing
- Datadog pricing list
- Datadog custom metrics billing
- SolarWinds Observability Self-Hosted pricing
- Amazon Managed Service for Prometheus docs
- Datadog data retention periods
- SolarWinds Observability endpoints docs
- Datadog OpenTelemetry docs
A cell without a source does not ship.
01The mechanics
What actually differs
The unit each option charges on, and who owns what afterwards, matter more than any single quoted figure.
Infrastructure monitoring watches the machines and networks everything else runs on: CPU, memory, disk, processes and network interfaces. It is the tier where the pricing question is sharpest, because the work is the most commoditized and every option measures much the same things. What differs is the unit you pay on, who holds the data, who owns the console, and whether the estate pays a license per node for collection that an open collector does without one.
The four options here are SolarWinds, the long-standing per-node monitor, SigNoz, an open source platform built on OpenTelemetry (the open standard for collecting this data), AMP (Amazon Managed Service for Prometheus) paired with Grafana dashboards, and Datadog's infrastructure tier. The table reads each vendor's own page, checked on 22 September 2026. For the dashboards on the AMP option, Amazon Managed Grafana charges $9 a month per active editor and $5 per active viewer in each workspace. The structural read holds whatever the list prices do: a per-node meter grows with the estate, a data meter grows with what you choose to send.
02The facts
Side by side, sourced
Every cell in this table carries the vendor's own page as its source and the date it was checked, and a cell without a source does not ship.
| SolarWinds | SigNoz | AMP with Grafana | Datadog Infrastructure | |
|---|---|---|---|---|
| What you pay on | Per node per month.SolarWinds Observability SaaS Network and Infrastructure Observability starts at $15.75 per node a month.SolarWinds Observability SaaS pricing, checked 2026-09-22 | Per GB of logs and traces and per million metric samples. No host-based and no user-based pricing.SigNoz Cloud charges $0.30 per GB of logs or traces and $0.10 per million metric samples, from $49 a month including $49 of usage.SigNoz pricing, checked 2026-09-22 | Per metric sample ingested, per GB stored, and per sample each query reads.Amazon Managed Service for Prometheus charges $0.10 per billion query samples processed.Amazon Managed Service for Prometheus pricing, checked 2026-09-22 | Per host per month.Datadog Infrastructure Pro is $15 per host a month billed annually, $18 on demand. Enterprise is $23 billed annually.Datadog pricing list, checked 2026-09-22 |
| What counts toward the bill | Every node that sent data in the previous day. A network device or host counts as one, a standalone virtual machine or cloud service as a third, a container as a tenth.SolarWinds Observability SaaS pricing, checked 2026-09-22 | The volume of data sent, however many hosts send it.SigNoz pricing, checked 2026-09-22 | Samples ingested and stored, and the samples dashboards and alerts read. Billed only when metrics are sent.Amazon Managed Service for Prometheus pricing, checked 2026-09-22 | Every host, each with an allotment of custom metrics pooled across the account.Datadog includes 100 custom metrics per host on Pro and 200 on Enterprise.Datadog custom metrics billing, checked 2026-09-22 |
| Commitment | A term subscription on a multi-year contract, billed annually. Subscription only.SolarWinds Observability Self-Hosted pricing, checked 2026-09-22 | SigNoz Cloud starts at $49 a month, which includes $49 of usage. The self-hosted community edition carries no license fee.SigNoz pricing, checked 2026-09-22 | No upfront fees or commitments. You pay only for what you use.Amazon Managed Service for Prometheus pricing, checked 2026-09-22 | Billed annually, month to month, or on demand. The annual rate is the lowest.Datadog pricing list, checked 2026-09-22 |
| Where it runs | Two products: Observability Self-Hosted, which you run yourself, and Observability SaaS, which SolarWinds runs.SolarWinds Observability Self-Hosted starts at $8 per node a month on Essentials.SolarWinds Observability Self-Hosted pricing, checked 2026-09-22 | Self-hosted in your own infrastructure, or SigNoz Cloud, a dedicated cloud, or managed by SigNoz inside your own cloud account.SigNoz pricing, checked 2026-09-22 | A serverless workspace in your AWS account, managed by AWS.Amazon Managed Service for Prometheus docs, checked 2026-09-22 | Datadog's cloud service, which Datadog runs.Datadog pricing list, checked 2026-09-22 |
| How long data is kept | Logs on Observability SaaS are kept from 1 to 60 days, chosen by plan. Longer retention goes through sales.SolarWinds Observability SaaS pricing, checked 2026-09-22 | Metrics on SigNoz Cloud are kept one month by default, with 3, 6 or 13 months available. Logs and traces start at 15 days.SigNoz pricing, checked 2026-09-22 | Metrics are kept 150 days by default, adjustable up to 1,095 days (three years).Amazon Managed Service for Prometheus docs, checked 2026-09-22 | Metric tags and values are kept for 15 months.Datadog data retention periods, checked 2026-09-22 |
| OpenTelemetry | SolarWinds Observability SaaS accepts OTLP (the OpenTelemetry Protocol) for logs, metrics and traces, from its own agent or any OpenTelemetry Collector.SolarWinds Observability endpoints docs, checked 2026-09-22 | Built on OpenTelemetry. Anything that can send OTLP can send to it.SigNoz pricing, checked 2026-09-22 | The OpenTelemetry Collector, in AWS's distribution, sends metrics to the workspace by Prometheus remote write.Amazon Managed Service for Prometheus docs, checked 2026-09-22 | Accepted through the Datadog Agent with its own OpenTelemetry Collector build, a standard OpenTelemetry Collector, or direct OTLP ingest.Datadog OpenTelemetry docs, checked 2026-09-22 |
03The honest row
When each is the right answer
Every option in this comparison is the right answer for somebody, and saying when is the part most comparisons leave out.
SolarWinds
The right answer when network device monitoring is the core need, the estate sits largely in its own data centers, and the operations team's fluency with the console carries real value that a rushed migration would burn. Self-Hosted keeps the whole system on your own ground.
SigNoz
The right answer for teams that want one OpenTelemetry-native tool across logs, metrics and traces, priced on data rather than hosts. Self-hosting the open source edition suits a team that can run it, and SigNoz Cloud suits one that would rather not, from a younger vendor than the others here.
AMP with Grafana
The right answer for AWS-heavy estates that want metrics in their own AWS account with no upfront commitment, and that already think in Prometheus. It is a metrics store, so logs and traces need a home of their own.
Datadog Infrastructure
The right answer when the estate already lives in Datadog and one console genuinely simplifies operations, on the hosts where its integrations and correlation with application data are used rather than merely billed.
04Questions
Asked about this comparison
Is replacing a per-node monitor risky?
The risk is managed by the parity bar: each alert and each rule is rebuilt on the open stack and proven to behave exactly as it does today, tier by tier and in writing, before a single host comes off per-node pricing. Until then both systems run side by side.
What does the open alternative actually consist of?
The OpenTelemetry Collector on every host, managed as a fleet over OpAMP (the open protocol for configuring collectors remotely), an open source time series database you own, and alerts and dashboards as code in your own repositories. No per-node or per-agent license sits anywhere in that path, and the console you watch it from is yours, running in your own account.
Does moving off a per-node monitor mean giving up the APM?
No. The ordinary host tier moves to the open stack, and the APM keeps the services where its depth earns its price. The collector can send the same host data to both while the parity bar is met.
Model it against your estate
The comparison that matters is the one run on your volumes and your contracts. The review reads them with you, and you leave with your version of the Logmetry Blueprint. No system access, no obligation.
Free. No system access. No obligation. You keep the drawing.
What you get
- Your architecture today, drawn
- The waste map
- Costs before and after, with estimates
- Your target architecture
- Options for your business problem
What it costsNothing