Sentry Continuous Profiling

Sentry Continuous Profiling helps development teams locate backend performance bottlenecks down to functions and lines of code, with a focus on Node.js and Python. Continuous profiles can run without a fixed time limit, and an Aggregate Flamegraph summarizes performance across a project without requiring one transaction as the scope. Developers can control profiling with start and stop APIs or configure it to begin when a root span is active. UI Profiling covers iOS, macOS, and Android; browser JavaScript and Electron are also listed. Sentry says profiling requires performance monitoring to be enabled in the app. Continuous and UI Profiling are billed by their respective profile-hour units and are available through pay-as-you-go usage. Sentry lists a free plan and paid plans from $26/mo. Integrations include GitHub, Jira, Slack, GitLab, Linear, Microsoft Teams, and PagerDuty. The GitHub integration can associate commits with releases and report unhandled issues in pull requests.

Sentry Continuous Profiling offers backend and UI profiling alongside project-wide flamegraph views and manual or trace-based controls. Consider the profile-hour billing and the requirement to enable performance monitoring before adopting it.

This tool suits developers investigating backend performance in Node.js or Python, as well as teams focused on frontend experience across mobile and browser platforms. It requires performance monitoring to be enabled in the app.

Sentry Continuous Profiling is a code-level performance tool for Node.js and Python services, with UI profiling for mobile and browser apps. It is best suited to teams already using Sentry performance monitoring across backend and client workloads. Its strongest advantage is combining long-running backend capture with project-wide flamegraphs; profile-hour billing and SDK-dependent availability call for a careful fit check.

Overview

Profiling helps teams trace performance bottlenecks to functions and lines of code, including backend slowdowns and infrastructure costs. Continuous profiles can run for arbitrarily long durations, making them useful when a short capture window would miss persistent or intermittent behavior. Profiling requires performance monitoring to be enabled in the app, so it is less compelling as a standalone tool.

Sentry is a cloud service hosted on Google Cloud Platform with a multi-tenant architecture. It states that its service has SOC 2 Type 2 and ISO 27001 certifications and a HIPAA attestation. Profiling support applies to SaaS customers, and continuous profiling availability depends on the SDK.

Compare the broader categories of Continuous Profiling Tools, Error Monitoring Software, Game Crash Reporting Software, and Mobile App Monitoring.

Key features

Backend and UI coverage

Continuous Profiling targets Node.js and Python backend code, while UI Profiling covers iOS, macOS, and Android. Browser JavaScript and Electron are also supported; browser UI profiling is listed for Angular, Ember, React, Solid, Svelte, and Vue SDKs. This breadth suits teams tracing performance across server and client code, but it does not remove the need to confirm that a specific SDK supports continuous profiling.

Project-wide analysis and controls

An Aggregate Flamegraph summarizes performance across a project without requiring one transaction as the scope. That makes it useful for looking at broader patterns rather than investigating only a single trace. The SDK's startProfiler and stopProfiler APIs allow manual control, while profileLifecycle can start profiling manually or when a root span becomes active. The profileSessionSampleRate option adds session-scoped sampling, giving teams a way to control capture.

Connected monitoring

Sentry lists integrations with GitHub, Jira, Slack, GitLab, Linear, Microsoft Teams, and PagerDuty. Its GitHub integration can link commits to releases, report unhandled issues in pull requests, and route assignments using a CODEOWNERS file. Frontend monitoring, release tracking, source-map support, and session replay are also supported, extending profiling into a wider monitoring workflow.

Pricing

Sentry uses a freemium model, with a free plan and paid access from $26/mo. The free plan includes a quota of 5000 events/mo, which gives smaller teams a way to start without a subscription but puts a ceiling on event volume. The free-plan retention period is 90 days.

Paid Continuous Profiling and UI Profiling are billed by Continuous Profile Hours and UI Profile Hours, respectively, through pay-as-you-go usage. The $26/mo entry price is not a flat measure of profiling cost: teams should account for their expected profile hours when assessing ongoing spend. No seat quota or trial length is stated. Sentry lists 30 days of data retention as well as a 90-day retention period; buyers should confirm which retention applies to the data and plan they need.

Platforms

Sentry lists Android, iOS, macOS, Linux, Windows, web, API, extensions, and self-hosted among its platforms. Its SDK coverage includes Node.js, Python, PHP, Browser JavaScript, Android Java/Kotlin, and iOS Swift/Objective-C, along with other languages and frameworks. The service is described as cloud-hosted, while self-hosted is also listed as a platform option.

Who it's for

This is a strong fit for teams that already enable Sentry performance monitoring and want backend profiling alongside mobile or browser UI profiling. Node.js and Python teams in particular can benefit from continuous capture without a 30-second duration limit, while the project-wide flamegraph helps when the question spans many transactions. Teams that need profiling without Sentry performance monitoring, or whose SDK does not support it, should look elsewhere.

Pros and cons

• Pros: Backend profiles can run without a fixed duration cap, giving teams more flexibility to investigate long-running behavior.
• Pros: Aggregate Flamegraphs show project-wide performance without tying analysis to one transaction.
• Pros: Manual APIs and trace-lifecycle controls let teams choose when profiling starts and stops.
• Cons: Profiling depends on performance monitoring being enabled, adding a prerequisite for teams seeking a focused profiling tool.
• Cons: Profile-hour billing means usage can affect cost beyond the advertised $26/mo starting price.
• Cons: Continuous profiling availability depends on SDK support, so platform coverage alone does not guarantee the needed profiling mode.

Alternatives

Choose Guance Container Monitoring if you want a free plan with stated host, metric-series, log-record, and rolling-retention allowances. Middleware is another freemium option, with a free plan that specifies data, RUM-session, synthetic-check, and browser-test allowances.

Alibaba Cloud ARMS RUM is an option for readers seeking a paid RUM offering without a free plan. Grafana Cloud Application Observability may suit teams comparing a free host-hour allowance with a usage-priced Pro plan.

LMbench is a free benchmark option for Linux and Windows. Consider OneUptime if an open-source core that can be inspected, modified, and self-hosted is a priority. Amazon CodeGuru Profiler offers a 90-day free tier before standard sampling-hour pricing applies. Coroot offers a self-hosted Apache 2.0 Community Edition alongside a paid Standard plan.

Verdict

Sentry Continuous Profiling is a good choice for teams already invested in Sentry that want long-running Node.js or Python profiling, project-level flamegraphs, and mobile or browser UI coverage in one monitoring workflow. Its combination of flexible capture controls and broad client coverage is the main reason to choose it. Look elsewhere if you need profiling independent of performance monitoring or want costs that are not tied to profile-hour usage.