asopi tech
asopi techIndie Developer

Observability Platform

Alopex OTel

Storage and dashboards in one product

Around OpenTelemetry, ingestion, storage, and visualization are usually separate products. The right composition differs by environment, with no fixed answer. Alopex OTel packages the whole set together. It scales from an embedded single process to multiple nodes while keeping the same screen and query language.

One package for the ingestion, storage, and visualization that otherwise surround OpenTelemetry.

ConceptOSS
Processes to start
1
Dashboard
Built in

Key points

  • Get ingestion, storage, and visualization together without selecting a combination
  • The dashboard is embedded in the same process rather than built separately
  • The UI and API stay the same as scale changes
observabilityopentelemetryrust

Challenge

A change in scale means rebuilding observability

A one-machine setup and a multi-node setup are different things. Products, screens, and query languages are all replaced. Alopex OTel aims to use the same screen and queries from embedded deployments through clusters.

  • Before starting, someone must decide what to combine and how
  • When scale changes, the deployment and dashboards must be rebuilt
  • Every migration also raises the question of what to do with stored data

Observability is usually built by combining ingestion, storage, and visualization. The composition differs by environment, and changing scale means choosing again.

Alopex OTel packages the entire set in one binary. Its dashboard is embedded in the same process. From an embedded single process to multiple nodes, the screen and query language remain the same.

Inside, Skulk handles metrics, Trail handles traces and logs, and Chirps acts as the logical router that absorbs differences in deployment scale. Chirps transparently routes in-process and network calls, so callers do not change when moving from embedded to distributed operation.

Add nodes while keeping one binary

Alopex OTel puts ingestion through visualization in one product. Start in an embedded single process, then move to a local server, two or three nodes, and multiple nodes without changing the screen, API, or query language.

  • One process to start and no wiring configuration
  • One location for retention policy
  • Nothing new to learn when scale changes
Processes to start
1
UI changes when scaling
None

Capabilities

Four scales

Change only the scale while keeping the same screen and query language.

Embedded

For single-process applications

  • Ship it inside an application without an external process
  • Keep observability entirely within the application

Local server

For development environments and small teams

  • Run it on one machine and send data from multiple applications
  • Keep the dashboard in the same process

Two or three nodes

For the point where redundancy becomes necessary

  • Expand by adding nodes
  • No need to reassemble the architecture

Multiple nodes

For production workloads

  • The screen and API stay unchanged as scale changes
  • No need to move stored data into another system

Features

Key features

What brings ingestion through visualization into one product.

Built-in dashboard

Visualization runs in the same process. There is no separate dashboard to construct and reconnect to each store.

Prometheus-compatible query API

Reuse existing queries and integrations, and keep the same language as requirements grow.

Metrics, traces, and logs in one place

Do not split storage by data type; follow them across the same interface.

One retention policy

Settings and retention periods live in one place. There is no separate configuration language per product.

Transparent from embedded to distributed

Move deployment scale without changing callers; in-process and network access work the same way.

Tech Stack

Technology stack

Combines components from the Alopex family.

Metrics

Alopex Skulk

Arrow + ParquetPrometheus Remote Writev0.3.0 released

Traces and logs

Alopex Trail

No schema required for writesRetains conflicting typesThe reader chooses the type

Routing

Chirps

SWIM + QUICMakes in-process and network access transparent

Status

Development status

The metrics path is running in Skulk v0.3.0 and accepts Prometheus Remote Write. The dashboard, traces and logs through Trail, and scale-out are next.

Public Resources

The latest documents, code, and community links in one place.

Get Involved

What are you combining today?

Knowing where ingestion, storage, or visualization creates work in real deployments helps decide what to build first.