July 2023 Summaries

Distributed tracing is a powerful technique for tracking the flow of requests through a system, providing valuable insights into performance and helping identify bottlenecks. It allows developers to link operations and requests between multiple services, offering a clear view of the entire operational flow of an action. This visibility enables pinpointing performance bottlenecks throughout the application stack, ultimately improving service quality and reliability. Distributed tracing is essential for applications built on microservices or serverless architecture, but its benefits also apply to traditional architectures. By understanding how distributed tracing operates behind the scenes, developers can harness this technique to optimize application performance and identify areas for improvement.

You've outgrown your mobile monitoring solution because, like a bike rider upgrading from a free but clunky bike to a better one, you need more context and data to identify and solve crashes and errors effectively. You're wasting time on broken workflows, ignoring alerts that are no longer meaningful, and questioning the accuracy of the data you're seeing due to discrepancies with other reporting tools. Furthermore, your users are complaining about issues, which can lead to frustration, bad reviews, and abandonment of your app. With these signs, it's clear that investing in a better tool is necessary to improve mobile stability and performance, gain higher ratings, and boost visibility on app stores.

New features in June's product updates for Sentry include the release of the new Codecov CLI, which allows users to easily configure code coverage settings and identify gaps in their test coverage directly from their terminal. Additionally, a Crons Timeline View has been added, providing an intuitive overview of all Cron Monitors' status within a specific timeframe. Java developers can now link their source code to stack traces using Sentry's new Java Source context feature, which includes automated setup through Gradle and Maven plugins. Furthermore, SQL queries are now displayed with improved readability in the Sentry platform.

The text discusses how modern concurrency APIs in mobile applications can help improve performance by detecting and offloading non-UI work from the main thread, which handles drawing the UI and responding to user actions. The main thread can become busy and delay subsequent UI renders if it's kept occupied with unrelated work, leading to issues such as dropped frames, janky animations, and unresponsive UIs. Sentry, a performance monitoring tool, can help identify these issues by detecting dropped frames and suggesting tasks that can be moved off the main thread. By offloading non-UI work to background threads, apps can maintain smooth UX and respond quickly to user interactions, making them more engaging and efficient.

The COVID-19 pandemic led to a significant surge in mobile usage, with daily time spent on mobile devices increasing by 39% in the US compared to three years ago and by 9% from late 2021. This growth forced developers to re-evaluate their tooling, particularly when it came to monitoring and fixing bugs, crashes, and performance issues. Companies like Delivery Hero and Relive saw rapid growth in users, with Delivery Hero's orders increasing almost doubling in Q1 2020, while Relive's user base grew from 3 million to over 15 million between March 2019 and November 2020. To address these challenges, both companies turned to Sentry for crash reporting and error monitoring solutions, which helped them improve their mobile experiences, reduce crashes, and increase overall app quality. As of June 2023, the number of mobile teams using Sentry has increased by 240%, with a significant rise in daily errors sent since January 2020.

Session Replay's overhead is a common concern, and its impact varies depending on a company's performance budgets. To measure this overhead, Sentry offers a web benchmark package, allowing users to evaluate the effect on their applications by capturing system resources and web-focused metrics, such as Google's Core Web Vitals. Session Replay records browser DOM states and mutations, with tasks like serialization and masking text content performed by the SDK. Sentry reduces main thread blocking by completing CPU-intensive tasks on a web worker. Memory usage varies with different sampling modes, and network traffic can be significant, primarily for mobile users with slower connections. Performance data collection is automated using Playwright, which simulates real user conditions and gathers metrics through the Performance API and Chrome DevTools Protocol. This allows developers to create test scenarios for their applications and evaluate performance impacts, ultimately providing quantifiable metrics for informed decision-making.