Observability Pipelines: 7 Tools compared

General-purpose telemetry pipelines that route, filter and transform logs, metrics and traces before they reach a backend. These tools carry security data but are not sold specifically for security operations; vendors positioning their…

7 tools

Quick comparison

All observability pipelines tools side by side, alphabetical.

ToolDeploymentPricing modelOpen sourceStandards / certs
Azure Data ExplorerCloudConsumption-based (compute + storage)
Datadog Observability PipelinesCloud + Self-hostedVolume-based (per GB processed)
FluentdSelf-hostedOpen sourceYes
MezmoCloudIngest-based (per GB)SOC 2 Type IIISO 27001HIPAA
SawmillsCloud + Self-hostedUsage-based (ingest volume; annual commitment)
Splunk Data Stream ProcessorCloudBundled with Splunk licensing
VectorSelf-hostedOpen sourceYes

Azure Data Explorer

Observability Pipelines
Best fit for

Microsoft-centric organizations wanting a scalable security data lake with powerful KQL analytics at lower cost than SIEM

Azure Data Explorer (ADX) is a fast, fully managed data analytics service from Microsoft designed for real-time analysis of large volumes of streaming data. While primarily a data analytics platform, ADX is increasingly used as a security data pipeline and lake for organizations that want to store, query, and analyze security telemetry at scale with Kusto Query Language (KQL), the same query language used by Microsoft Sentinel.

Pricing

Pay-as-you-go (compute + storage) / Reserved capacity discounts

Consumption-based (compute + storage)

Deployment

Cloud

Datadog Observability Pipelines

Observability Pipelines
Best fit for

Organizations already using Datadog that want managed pipeline capabilities with enterprise support and monitoring

Datadog Observability Pipelines is a data routing and transformation tool that allows organizations to collect, transform, and route observability data from any source to any destination. Built on the open-source Vector project, it provides a managed pipeline experience with Datadog's enterprise support, monitoring, and integration ecosystem. It helps teams control data volumes and costs while maintaining visibility across their infrastructure.

Pricing

From $0.10/GB processed / Enterprise custom

Volume-based (per GB processed)

Deployment

CloudSelf-Hosted

Fluentd

Observability Pipelines
Best fit for

Cloud-native teams wanting a lightweight, proven open-source data collector with a massive plugin ecosystem

Fluentd is an open-source data collector that unifies data collection and consumption for better use and understanding of data. A CNCF graduated project, Fluentd provides a unified logging layer that allows you to collect data from multiple sources, transform it, and route it to various destinations. Its plugin-based architecture and lightweight footprint make it a popular choice for containerized and cloud-native environments.

Pricing

Free (open source) / Commercial support via vendors

Open source

Deployment

Self-HostedOpen Source

Mezmo

Observability Pipelines
Best fit for

Teams wanting combined log management and pipeline capabilities with a developer-friendly experience

Mezmo (formerly LogDNA) is a log management and observability pipeline platform that helps teams collect, process, route, and analyze log data at scale. Its Telemetry Pipeline product provides real-time data routing and transformation capabilities, enabling organizations to control where their data goes and how it is shaped before reaching downstream destinations like SIEMs, data lakes, and monitoring tools.

Pricing

From $0.80/GB ingested / Enterprise custom

Ingest-based (per GB)

Deployment

Cloud

Standards & certifications

SOC 2 Type IIISO 27001HIPAAPCI DSS

Sawmills

Observability Pipelines
Best fit for

Teams on Datadog, Splunk or similar that want to cut observability cost by filtering and optimising telemetry before ingestion

Sawmills is an AI-driven (agentic) telemetry pipeline that sits upstream of your observability backend and processes logs, metrics and traces in real time before they are ingested. It filters, transforms and improves the quality of telemetry pre-ingestion to cut observability spend, and is built on the OpenTelemetry Collector so it works alongside tools like Datadog and Splunk rather than replacing them.

Pricing

Usage-based on ingest volume with an annual commitment. Public pricing lists from $2,500/month (2026) with a free trial; verify current pricing with the vendor.

Usage-based (ingest volume; annual commitment)

Deployment

CloudSelf-Hosted

Splunk Data Stream Processor

Observability Pipelines
Best fit for

Existing Splunk customers wanting to optimize data flows and reduce ingest costs within the Splunk ecosystem

Splunk Data Stream Processor (DSP) is Splunk's real-time stream processing engine designed to collect, process, and deliver data at scale. Built on Apache Flink, DSP enables organizations to filter, mask, enrich, and route data in real time before it reaches Splunk or other destinations. It is positioned as a complement to Splunk Enterprise for organizations that need to optimize data flows and reduce ingest costs.

Pricing

Included with Splunk Cloud / Enterprise add-on pricing

Bundled with Splunk licensing

Deployment

Cloud

Vector

Observability Pipelines
Best fit for

Teams wanting the highest-performance open-source pipeline with Rust-based reliability for high-throughput data routing

Vector is a high-performance, open-source observability data pipeline built in Rust. Originally created by Timber.io and now maintained by Datadog, Vector collects, transforms, and routes all log, metric, and trace data with a focus on reliability and performance. Its Rust-based architecture delivers significantly better performance than alternatives written in higher-level languages, making it ideal for high-throughput environments.

Pricing

Free (open source, MPL 2.0)

Open source

Deployment

Self-HostedOpen Source

Related guides

Other categories you might be evaluating alongside observability pipelines.

About this listing

Observability Pipelines tools, listed alphabetically and compared on public information. How we work →

Frequently Asked Questions

Both sit between your data sources and your backends, filtering and routing telemetry to control cost. The difference is what the vendor builds and sells for. Observability pipelines handle logs, metrics and traces for engineering and reliability teams. Security data pipelines are built around SOC workflows: normalising security telemetry, preserving detection quality, and feeding SIEM and detection tooling. Several observability pipelines are used successfully for security data, and the two categories overlap in practice.

Yes, and many teams do. Filtering low-value logs before ingestion cuts licensing spend regardless of which pipeline does it. What you give up is the security-specific handling: parsers for security formats, awareness of which fields detections depend on, and threat intelligence enrichment. If SIEM cost is the only goal, an observability pipeline may be enough; if detection quality matters, compare against the security data pipeline category.