Skip to main content

xScaler Managed Metrics Backend

xScaler Managed Metrics Backend is a fully managed, Prometheus-compatible metrics backend built for teams that need reliable, scalable observability without the overhead of running infrastructure themselves.

What it is​

xScaler acts as a drop-in remote storage target for any Prometheus-compatible stack. You push metrics from your existing Prometheus, Grafana Alloy, OpenTelemetry Collector, or OTel SDK — xScaler handles ingestion, storage, and querying at scale.

  • Prometheus-compatible — works with any tool that speaks Prometheus remote_write or PromQL.
  • Multi-tenant — each tenant's data is fully isolated via the X-Scope-OrgID header.
  • Fully managed — no servers, no WAL, no compaction jobs to maintain.

Architecture​

Your instrumented application or infrastructure emits metrics through an OTel SDK, Prometheus, or Grafana Alloy. Those metrics flow into the xScaler backend over remote_write or OTLP. Grafana (or any PromQL-capable tool) queries the data back out.

Supported ingest methods​

MethodProtocolEndpoint
Prometheus remote_writeHTTPPOST /api/v1/push
OpenTelemetry CollectorOTLP/HTTPPOST /otlp/v1/metrics
OpenTelemetry CollectorOTLP/gRPC:443 (TLS)
Grafana AlloyHTTP (remote_write)POST /api/v1/push
OTel SDK (Python / Node.js / Go)OTLP/HTTPPOST /otlp/v1/metrics

Supported query methods​

xScaler exposes the full Prometheus HTTP API at /prometheus/api/v1/..., so any tool that supports PromQL works — Grafana, Prometheus-compatible dashboards, custom scripts using curl, and more.

Multi-tenancy​

Every request to xScaler — both writes and reads — must include two HTTP headers:

Authorization: Bearer <token>
X-Scope-OrgID: <tenant-id>

The X-Scope-OrgID header is the tenant isolation header. Without it, xScaler cannot identify which tenant namespace to write into or query from, and the request will be rejected with a 400 or 401 error.

info

These two headers are mandatory on every single request. See Authentication for full details.

Next steps​