March 2026 Summaries

OpenTelemetry has introduced profiling as the fourth signal within its project, marking its protocol's entry into the Alpha stage, with the OpenTelemetry Profiling Working Group playing a key role in this development. The post highlights how to use the opentelemetry-ebpf-profiler with an OpenTelemetry collector distribution, which can also be integrated into the Polar Signals Cloud for profiling data analysis. It provides specific configurations for utilizing the profiler alongside Polar Signals Cloud and the open-source Parca server, demonstrating practical applications of the profiling capability. While currently only JIT, interpreted, or Go stacks are symbolized on-host, the group plans to expand support for asynchronous symbolization and uploading symbol data to a backend. The post concludes by acknowledging the collective efforts that have contributed to this advancement and invites interested individuals to join the ongoing work in the OpenTelemetry Profiling Working Group.

The text discusses the mechanisms of CPU profiling, focusing on hardware timers and eBPF technology. It explains how modern operating systems use hardware timers to facilitate non-cooperative multitasking, allowing multiple tasks to appear as if running simultaneously by periodically halting the CPU to switch tasks. Profiling utilizes these timers to measure elapsed time, employing Linux's perf events to register events that trigger code execution at specific intervals. Traditional profilers like perf rely on frame-pointer-based unwinding or DWARF-based methods, which can be resource-intensive and limited by context switching. However, the introduction of eBPF, an extension of the classic BPF initially used for network analysis, allows for more efficient profiling by enabling user-mode programs to attach custom logic to kernel events without context switching. eBPF programs are verified for safety and run directly in the kernel, making them capable of unwinding stacks and passing data to user-mode profilers like Parca without the overhead of copying data between processes. This approach enhances performance while maintaining system stability, as eBPF is not Turing-complete and cannot crash the kernel. The text concludes by highlighting the collaboration of hardware timers and eBPF in modern CPU profilers, supported by an AI usage disclaimer about the article's preparation.

The VSCode Polar Signals Profiler extension integrates profiling data directly into the code editor, allowing developers to view CPU time, memory allocations, and other performance metrics as inline annotations alongside their source code. This tool supports both Polar Signals Cloud and self-hosted Parca instances, and is compatible with any editor that supports the VS Code extension API via the Open VSX Registry. Users can connect to their profiling data through OAuth for Polar Signals Cloud or directly to a Parca instance without authentication. The extension offers built-in presets for common profiling scenarios and allows custom queries for more specific investigations, enhancing workflows with AI integration features that facilitate data sharing with AI assistants. Installation is straightforward, and the tool is open source, inviting contributions and feedback from the community to improve its functionality.