Uptrace

What makes Uptrace different

Uptrace distinguishes itself by being an OpenTelemetry-native observability platform that unifies traces, metrics, and logs into a single interface, eliminating the need for fragmented toolchains. Built on ClickHouse, a high-performance columnar database, it processes billions of spans at a fraction of the cost of legacy providers like Datadog or New Relic. The architecture leverages the OTel Arrow protocol for ingestion, achieving 15–30x compression compared to raw telemetry data, which significantly reduces egress fees and storage costs.

Unlike traditional SaaS observability tools that charge per-seat or per-host, Uptrace bills based on data volume (ingested spans/logs) and metrics throughput. This model is particularly fair for microservices architectures that generate many small spans. The platform offers zero-code auto-instrumentation via battle-tested SDKs, allowing developers to get their first trace in under 10 minutes without vendor lock-in, as the instrumentation code remains standard OpenTelemetry.

Uptrace supports a “cloud-only” managed service for teams seeking zero ops, as well as self-hosted and on-premises options for strict compliance requirements. The managed cloud service is hosted in EU data centers (Germany and Finland), ensuring GDPR compliance and keeping data within the European Union. It provides a 99.9% uptime SLA and 16/5 enterprise support, bridging the gap between open-source flexibility and enterprise reliability.

Pricing model

Uptrace uses a transparent, usage-based pricing model with no per-seat fees. The free tier includes 1TB of span data and 100k metrics per month. Paid plans scale with data ingestion and timeseries volume:

• Spans/Logs: Billed per GB ingested. Standard retention is 4 weeks. Cold storage (S3/HDD) for data beyond 4 weeks costs $0.01 per GB-month. Dropped data (sampling) costs $0.01 per GB.
• Metrics: Billed per million ingested datapoints. The rate starts at $0.025 per 1M datapoints and decreases with volume. A timeseries is defined as a metric with unique labels.
• Infra Hosts: Some plans bundle infrastructure monitoring capacity (e.g., 350 hosts).

This model allows users to pay only for what they store, avoiding the hidden costs of per-host licensing or expensive egress fees associated with other platforms.

When it fits

• OpenTelemetry Adopters: Teams already using or planning to adopt OpenTelemetry who want a backend that natively supports it without complex adapters.
• Cost-Conscious Engineering Teams: Organizations looking to reduce observability bills by up to 90% compared to Datadog or New Relic, especially those with high-volume telemetry data.
• EU-Centric Companies: Businesses requiring strict GDPR compliance and data residency within the European Union.
• Microservices Architectures: Environments with many small services generating numerous spans, where per-span or per-host pricing would be prohibitive.

When it doesn’t

• Non-OpenTelemetry Shops: Teams heavily invested in proprietary agent-based solutions (like Datadog Agent) that do not easily export to OTel may face migration friction.
• Global Multi-Region Needs: Currently, Uptrace Cloud is limited to EU data centers; teams requiring native low-latency ingestion in the US or Asia may need to self-host or accept higher network latency.

Inclusion criteria

Uptrace meets all three inclusion criteria:

1. Transparent Pricing: Detailed pricing per GB and per datapoint is published on their website.
2. Self-Service Signup: Users can start a free trial or use the public live demo without sales intervention.
3. Public SLA/Status: A 99.9% uptime SLA is stated for the Cloud plan, and a community-supported status is implied through their public deployment options.