Skip to main content

FAQ

Common questions about what StereumLabs is, how it works, and how to use it. If your question isn't covered, write to contact@stereumlabs.com.

About StereumLabs​

What is StereumLabs?​

A measurement and observability platform for Ethereum execution and consensus clients. We run all major EC and CC implementations side-by-side on declared hardware (bare-metal in Vienna, plus GCP cloud), publish neutral comparisons, and expose the underlying dashboards, metrics, and logs to subscribers. See Introduction.

Who operates StereumLabs?​

RockLogic GmbH, an Austrian Ethereum infrastructure company. The data-science and dashboard authorship is done in collaboration with MigaLabs. The project is supported by an Ethereum Foundation grant.

How are you different from other Ethereum monitoring tools?​

Most tools focus on your node: alerts, uptime, your validators' performance. We focus on the client implementations themselves. We compare them under identical conditions on identical hardware, across every supported version, so operators and researchers can make informed decisions about which client to run and when to upgrade. Our data is not for incident response on your own infrastructure. It's a reference baseline.

Is StereumLabs neutral, or do you favor a specific client?​

Neutrality is a foundational goal. We use equalized scenarios, identical resource budgets, identical sampling, and document any non-default configuration with the client team's recommendation. See the Neutral methodology section. If you spot what looks like a methodological bias, please flag it.

How do I cite StereumLabs?​

See Citing StereumLabs and For Researchers. For DOI or archival mirrors of specific datasets, contact us.

Access & plans​

How do I get access?​

Sign up via the StereumLabs dashboard (see Get Started). Free tier is self-serve. Pro and Enterprise require completing payment or contacting us.

What does it cost?​

See Plans & Prices. Free is EUR 0. Pro is EUR 3,200/month billed annually. Enterprise is custom.

I'm a client team developer. Can I get free access?​

Yes. Client development teams get free dashboard access for the duration of their engagement. Email contact@stereumlabs.com with your team affiliation. See For Client Teams.

I'm a researcher. Can I get free access?​

Independent researchers can request free Pro-tier access for the duration of an active research engagement. Email us with your affiliation and intended use. See For Researchers.

Can I share my account with my team?​

The Grafana user count per plan determines how many simultaneous accounts you can create. Free includes 1 user, Pro 5, Enterprise custom. See Plans & Prices.

Data​

How fresh is the data?​

  • Free tier: 7-day delayed recording rules only.
  • Pro and Enterprise: near real-time. Scrape interval is 15 seconds. End-to-end latency to queryable Prometheus is typically 25 to 30 seconds.

How long do you keep historical data?​

  • Free tier: 90 days.
  • Pro and Enterprise: unlimited (subject to ongoing storage capacity).

Can I export the data?​

Raw exports (CSV or Prometheus snapshots) are an Enterprise feature. See Plans & Prices. For citation-grade dataset snapshots, contact us.

Can I run my own queries?​

Yes. Pro and Enterprise users can write arbitrary PromQL against the full Prometheus datasource in their private Grafana org. Pro and Enterprise users also get Lucene queries against Elasticsearch (logs). See Build your own dashboards for the datasource names, custom labels, and PromQL examples.

What about logs?​

Container logs from every EC and CC instance are shipped via Filebeat to Elasticsearch. Both INFO and DEBUG levels are available, with DEBUG retained for a shorter window than INFO. Available on Pro and Enterprise tiers. Queryable from Grafana.

Where does the data physically live?​

Metrics and logs are stored in our infrastructure at the NDC2 datacenter in Vienna, Austria. RockLogic is ISO 27001 certified (conformity statement) and StereumLabs operations fall under that certification.

Methodology​

What clients do you cover?​

Execution: Besu, Erigon, Ethrex, Geth, Nethermind, Reth. Consensus: Grandine, Lighthouse, Lodestar, Nimbus, Prysm, Teku. Plus the standalone Caplin (Erigon's built-in CL).

What pairings do you run?​

All 36 EC×CC combinations on bare metal in Vienna (NDC2), six supernode pairings on GCP, plus one Caplin standalone. 43 active configurations at the time of writing. See List of VMs.

How do you handle client-specific configuration?​

We start from each client's defaults and only apply documented, client-team-recommended flags required for correctness or scenario fitness. Any deviation is justified in the methodology notes for that run. See Neutral methodology.

What about staking metrics?​

Validator clients are connected on the CC side of our fleet, so blob and block-proposal behavior is realistic. Staking-focused dashboards (participation, inclusion delay, missed duties) are still under development. See Roadmap.

What is a supernode?​

A consensus-layer node that custodies all 128 PeerDAS data columns instead of the default 4 custody groups. See the Glossary. We label supernodes with cc_client="<name>-super".

Why bare-metal and cloud?​

Bare-metal is the controlled, low-noise reproducibility anchor. Cloud (currently GCP) reflects the reality most operators face: variable hardware, noisy neighbors, default firewalls. We measure both so users can map our numbers to their own environment.

Why are the absolute resource numbers higher than what I see at home?​

Our published numbers come from a server-class fleet (Epyc 9654P, NVMe, dedicated NICs). Home setups (NUC, mini-PC, consumer SSDs) will see different absolute numbers but the relative comparisons (which client uses less memory, which syncs faster) carry over. See the note in For Home Stakers.

How do you handle hardforks?​

For each hardfork:

  1. Upgrade fleet to fork-compatible versions before the activation slot.
  2. Take 14-day before/after windows for fleet-wide comparisons.
  3. Publish a post-fork analysis on the blog (e.g., Fusaka hardware impact).
  4. Update the dashboards and metrics catalog with any new fork-specific panels.

Operations​

My client or version isn't covered. Can you add it?​

For supported clients, new versions are picked up as part of the normal fleet update cycle. For new client implementations, email us. Adding a new client typically requires a few weeks of fleet work and stabilization.

I want a custom metric scraped or panel added. How?​

Email contact@stereumlabs.com with the metric name (or panel definition) and a one-line description. We review additions in the normal weekly cadence.

Can I integrate this into my own Grafana?​

Pro and Enterprise users get a Grafana account in our managed instance with their own private organization. Self-hosted Grafana with our datasources as a backend is an Enterprise-only scenario. Contact us.

Is there an API?​

Direct programmatic API access is an Enterprise-only feature, tailored individually to the customer's integration needs. On Free and Pro, the datasources are reachable through Grafana with standard Grafana auth. There is no separate REST or GraphQL layer on top.

Is there an SLA?​

See Support & SLA for support channels and response expectations.

A status page?​

A public uptime and incident status page is on the Roadmap. For incident questions in the meantime, contact us.

Practical​

I'm new. Where should I start?​

Start with the Start here section and pick the page that matches your role:

If you'd rather start from the data: the blog is the most opinionated entry point (Execution Client Sync Speed, EC P2P peering, Caplin standalone are good starters). For methodology, read Purpose & Scope.

I want to compare clients before picking one. Where do I start?​

The blog is the most opinionated entry point. Every post is a head-to-head comparison under identical conditions. The dashboard catalog gives you the underlying data. The comparator dashboards put two configurations side-by-side.

What if I find a bug in a measurement or a wrong number on a dashboard?​

Email contact@stereumlabs.com with the dashboard URL, blog link, or panel name and what looks wrong. Material corrections are logged in the Changelog.

Change control for this page: material edits will be logged in the global Changelog with a short rationale and effective date.