Our Blog

Welcome to the Polar Signals blog where we write about profiling, performance, and company culture.

Introducing the Sandwich View

A new way to visualize a function’s callers and callees side by side.

January 8, 2026 | Yomi Eluwande

Data Visualization

Frontend

Profiling

Caller / Callee View

Sandwich View

Optimizing SLO Performance at Scale: How We Solved Pyrra’s Query Performance on Thanos

Learn how we optimized Pyrra’s SLO monitoring to handle high-cardinality metrics at scale, reducing cross-zone traffic and query times while maintaining 99% accuracy.

December 30, 2025 | Matthias Loibl

Performance

Observability

SLOs

Prometheus

Thanos

Profiling NVIDIA CUDA in Kubernetes

The easiest way to get started Profiling CUDA in Kubernetes

December 18, 2025 | Frederic Branczyk

CUDA

NVIDIA

Profiling

GPU

Kubernetes

USDT Probes: A Deep Dive

How we added USDT Probe support to Parca Agent

December 10, 2025 | Tommy Reilly

eBPF

Tracing

USDT

Great Lakes: A Delta Lake Fork

Our updates to Delta Lake Table Format

December 4, 2025 | Thor Hansen

Column Database

Delta Lake

Databases

Questioning an Interface: From Parquet to Vortex

Breaking free from the shackles of interface-imposed performance limitations

November 25, 2025 | Alfonso Subiotto Marqués

Databases

Datafusion

Vortex

Apache Parquet

Custom Labels for Node.js

Correlate Application-Specific Data With Profiles

November 19, 2025 | Brennan Vincent

Correlation

Javascript

Profiling

Continuous NVIDIA CUDA Profiling In Production

Low-Overhead Profiling of GPUs with USDT Probes and eBPF

October 22, 2025 | Tommy Reilly

GPU

Profiling