Skip to content
Telopi
Book a demo

Hosted in Germany

Your telemetry. Our operations.

Telopi gives you your own Grafana, with Mimir, Loki and Tempo behind it, hosted and operated for you in Germany. Send metrics, logs and traces with Grafana Alloy, or have your website’s uptime checked with nothing to install.

Book a demoOpen the live demo

The live demo needs no account.

grafana / dashboards / service-overviewSample data
Range
SLO target
service = checkout-apiPAUSED

Traffic

312req/s

Errors

0.16%

Latency (p95)

82ms

Saturation

11.7in-flight

Latency percentiles

golden-signals · Tempo span metrics
p95p50
Latency percentiles over the selected range, in milliseconds
Timep50p95
23h 45m ago34.188.8
23h 30m ago34.578.2
23h 15m ago34.587
23h ago3692.9
22h 45m ago32.277
22h 30m ago35.582.7
22h 15m ago36.786.2
22h ago36.693.6
21h 45m ago35.783
21h 30m ago35.779
21h 15m ago32.875.9
21h ago34.887.6
20h 45m ago35.391.5
20h 30m ago33.684
20h 15m ago37.595.7
20h ago35.382.3
19h 45m ago3679.4
19h 30m ago34.579.6
19h 15m ago35.384
19h ago39.791.8
18h 45m ago37.995.6
18h 30m ago3994.1
18h 15m ago38.393.9
18h ago40.690.8
17h 45m ago37.690.8
17h 30m ago38.490
17h 15m ago40.995.4
17h ago41.292.1
16h 45m ago39.494.2
16h 30m ago42.698.1
16h 15m ago41.299.8
16h ago42.598
15h 45m ago41.7106.1
15h 30m ago39.999.5
15h 15m ago44.2109.2
15h ago41.5104.6
14h 45m ago44.9111.2
14h 30m ago42.2105.4
14h 15m ago43.5105.7
14h ago44.7105.6
13h 45m ago41.7106.7
13h 30m ago41.4103.1
13h 15m ago45.8103.7
13h ago43110.1
12h 45m ago45.7106.6
12h 30m ago42.8104.4
12h 15m ago42.8103
12h ago42.4101.6
11h 45m ago43.9100.8
11h 30m ago44.4107.6
11h 15m ago41.598
11h ago43.298.2
10h 45m ago42.999.5
10h 30m ago41.8101.4
10h 15m ago46.3106.7
10h ago45.1108.2
9h 45m ago45.4121.5
9h 30m ago47.2126.6
9h 15m ago48.2131.5
9h ago47.3147.9
8h 45m ago52.2176.6
8h 30m ago53.4191.3
8h 15m ago55221.5
8h ago57.6239.4
7h 45m ago59.7265.4
7h 30m ago58260.7
7h 15m ago58.4262.6
7h ago54.4241.7
6h 45m ago55.2224.3
6h 30m ago48.6192.4
6h 15m ago47.6162.1
6h ago45.3148.9
5h 45m ago44.6127.9
5h 30m ago42.6111.4
5h 15m ago40.9107.9
5h ago38.8102.7
4h 45m ago38.395.5
4h 30m ago38.896.8
4h 15m ago3991
4h ago34.584.2
3h 45m ago38.688.5
3h 30m ago36.688.1
3h 15m ago36.684.6
3h ago33.874.9
2h 45m ago36.489.4
2h 30m ago35.880.5
2h 15m ago32.471.8
2h ago32.474.8
1h 45m ago3587.1
1h 30m ago34.483.6
1h 15m ago34.686.4
1h ago34.184.3
45m ago33.674.5
30m ago33.685.3
15m ago33.787.5
now35.981.9

Error budget

slo-availability
10%budget left
Error budget remaining
10.0%
Latency objective met
98.2%
Availability vs objective
99.91% / 99.9%

Traffic by operation

golden-signals · now
  • GET /api/orders106 req/s
  • GET /api/products/:id81 req/s
  • POST /api/checkout50 req/s
  • POST /api/cart41 req/s
  • GET /healthz34 req/s

Nothing to instrument

Every panel here is computed from Tempo's span metrics. A service shows up the moment it sends OTLP traces — no metrics instrumentation, no recording rules, no dashboard JSON to maintain.

Golden signals and SLOs, before you build a panel.
Integrations

Built on the stack you already trust

Grafana, Prometheus, Mimir, Loki, Tempo, Alloy, OpenTelemetry and NVIDIA DCGM, running on AWS, AWS European Sovereign Cloud, Azure, StackIT, Hetzner, Kubernetes or Docker Compose.

Observability stack

  • Grafana
  • Prometheus
  • Mimir
  • Loki
  • Tempo
  • Alloy
  • OpenTelemetry
  • NVIDIA DCGM

Clouds and runtimes

  • AWS
  • AWS European Sovereign Cloud
  • Azure
  • StackIT
  • Hetzner
  • Kubernetes
  • Docker Compose
Platform

One control plane. Infrastructure you own.

Connect a cloud account, provision an environment, and manage its config, version and health from one portal. The stack runs inside your infrastructure, next to your workloads — including your GPU nodes.

Sample dataLIVE

Your cloud account

AWS · Azure · StackIT

GPU nodes

DCGM · NVIDIA

Telopi control plane

Provisioning · lifecycle

Grafana stack

Metrics · logs · traces

Telemetry store

12ms p50

Optiop agent

Roadmap

GPU utilization

71%

Tokens/s

18.4k

TTFT p95

340ms

Telemetry egress to Optiop

0bytes

Agentic SRERoadmap

Agents that will read your dashboards
before your on-call does.

As Telopi grows into a full SRE platform, Optiop operators will correlate telemetry, runbooks, and incident history to investigate anomalies and recommend remediation directly inside your Grafana workspace.

system.logstreaming
›environment prod-eu1 · health check passed (204ms)
›provisioning grafana-instance-04 on aws/eu-west-1 ...
›config sync: dashboards updated (v11.4.2)
›ticket #482 opened: "increase retention to 30d"
›instance stage-us1 status: running

Secure by default

Enterprise-grade isolation, encryption at rest, and least-privilege access baked into every provisioned workspace.

Live telemetry, zero setup

Metrics, logs, and traces flow into a managed Grafana instance the moment you connect a data source.

Bring your own cloud

Provisioned via provider-specific IaC into your AWS, Azure, or StackIT account — your infrastructure and telemetry data never leave your control.

Scale without the scramble

Elastic infrastructure that meets demand before your dashboards notice a spike.

How it works

From first call to dashboards in three steps.

No servers to run and no stack to upgrade. You send telemetry; we keep everything behind it running.

  1. Get your own Grafana

    We set up your tenant: a Grafana of your own, an ingest endpoint and the credentials to send with. Your data is kept apart from every other customer’s.

  2. Send your telemetry

    Point Grafana Alloy at your endpoint, or send OTLP traces directly. For a website, a short snippet is enough.

  3. See it in Grafana

    Dashboards, alerting and Explore over your metrics, logs and traces, in a Grafana that is yours alone.

config.alloy
prometheus.remote_write "telopi" {
  endpoint {
    url = "https://ingest.telopi.com/acme/mimir/api/v1/push"

    basic_auth {
      username = "acme"
      password = sys.env("GAAS_PASSWORD")
    }
  }
}

prometheus.exporter.unix "host" { }

prometheus.scrape "host" {
  targets    = prometheus.exporter.unix.host.targets
  forward_to = [prometheus.remote_write.telopi.receiver]
}
Grafana Alloy, sending metrics to the tenant “acme”.

Security & data residency

Your data is stored in Germany, and stays yours.

Telemetry says a lot about a business. We keep it where you can account for it, and apart from everyone else’s.

  • Stored in Germany

    The stack runs on servers in Nuremberg, and your metrics, logs and traces are stored there. Traffic passes through a CDN (Amazon CloudFront) that ends TLS at its edge locations.

  • Isolated per customer

    Every customer is a separate tenant, with its own Grafana, credentials and limits. Our gateway ties every request to the tenant’s ID, and we test that one tenant cannot read another’s data.

  • Encrypted in transit

    TLS from your agent to our ingest, and from your browser to your Grafana.

  • Open source underneath

    Grafana, Mimir, Loki and Tempo: standard formats, so you can move to another stack.

  • A German company

    Telopi is run by Optiop GmbH. We process data under German law and the GDPR.

Need it inside your own cloud account?

On request

Today Telopi runs on our servers in Nuremberg. If you need the stack in your own cloud account instead, tell us, and we will plan it with you.

Book a demo to discuss it
Where we’re headed

From provisioning
to agentic SRE.

Telopi starts as the control plane for observability infrastructure and grows into a full SRE platform — reliability engineering that gets more automated at every stage.

01Live

Observability Provisioning

Connect a cloud account and provision a Grafana stack with a few clicks.

02In progress

Managed Observability for AI workloads

Health, versions and provisioning across every environment, plus GPU and LLM dashboards.

03Next

SRE Platform

SLIs, SLOs, error budgets, alerting, incident management, and runbooks.

04Vision

Agentic SRE

AI agents investigate, diagnose and recommend — then execute approved fixes, including for GPU fleets: idle-GPU detection, failing-node isolation and capacity recommendations.

Support

When something needs
a human, we’re here.

Provisioning and lifecycle management are self-serve — but operational requests and incidents are one ticket away.

In-portal ticketing

Create and track support tickets from the customer portal, scoped to the environment they affect.

SLA-backed response

Priority tiers with committed response times on Business and Enterprise plans, from first byte to resolution.

Docs, status & runbooks

Setup guides and a live status page today — shared runbooks and escalation paths arrive with the SRE Platform.

FAQ

Questions, answered.

A managed Grafana LGTM stack: your own Grafana, with Mimir for metrics, Loki for logs and Tempo for traces, hosted and operated by us in Germany. You send telemetry; we run and update everything behind it.

With Grafana Alloy, or with anything that speaks OpenTelemetry (OTLP over HTTP), Prometheus remote write or the Loki push API. For websites, the Grafana Faro snippet. Each customer gets an ingest endpoint and credentials.

On servers in Nuremberg, Germany, operated by Optiop GmbH. Each customer is a separate tenant: your data is stored under your tenant ID, and other tenants cannot read it.

It depends on how much you send and how long you keep it. Book a demo and we will size it with you; the live demo costs nothing and needs no account.

Yes. Open the live demo to look around a real Telopi Grafana right away. After a demo call, we set up a tenant for you to try with your own data.

You keep your agents, configuration and dashboards: they are standard Grafana, Prometheus, Loki and OpenTelemetry, and work with any other LGTM stack.

Self-service sign-up, a usage and cost page, and rules to drop the metrics you do not need.

See your own telemetry in Telopi.

Book a 30-minute demo, or look around the live one first.

Book a demoOpen the live demo