Frederic Branczyk
@brancz.com
Founder polarsignals.com 🧊 ❄️ Building parca.dev, Prometheus maintainer. Distributed systems, databases, and performance engineering. Actually born in Berlin. (he/him)
I just realized that the feedback loop we created for our new database is its real superpower. Since *everything*, including the catalog, is on object storage, it allows us to validate/try out things super fast because everything can act independently against object storage.
November 26, 2025 at 2:08 PM
I just realized that the feedback loop we created for our new database is its real superpower. Since *everything*, including the catalog, is on object storage, it allows us to validate/try out things super fast because everything can act independently against object storage.
Just the move from parquet to vortex gave us a massive perf boost. But more importantly, it opens up the path for way more. So much low hanging fruit now!
We completed a major project to switch our storage file format from Parquet to Vortex 🌪️ resulting in 70% average query performance improvement across the board 🚀
Learn more about how rethinking interface-imposed limitations unlocked these gains in our latest blog post 👇
Learn more about how rethinking interface-imposed limitations unlocked these gains in our latest blog post 👇
November 25, 2025 at 3:08 PM
Just the move from parquet to vortex gave us a massive perf boost. But more importantly, it opens up the path for way more. So much low hanging fruit now!
One of our most popular features, now available with nodejs! Also a sneaky customer reference 👀
Ever wanted to filter flame graphs by distributed trace ID? Or by anything else? In Node.js, now you can! www.polarsignals.com/blog/posts/2...
Happy Profiling! 🚀🐻❄️
Happy Profiling! 🚀🐻❄️
Custom Labels for Node.js
Correlate Application-Specific Data With Profiles
www.polarsignals.com
November 19, 2025 at 5:58 PM
One of our most popular features, now available with nodejs! Also a sneaky customer reference 👀
It's so awesome that we can use our own product every single day at Polar Signals to understand our own software. I want to get into the rhythm and do more of these.
What do you think? How would you have troubleshooted this, and could you have understood what's going on in less than 1min?
What do you think? How would you have troubleshooted this, and could you have understood what's going on in less than 1min?
November 19, 2025 at 2:04 PM
It's so awesome that we can use our own product every single day at Polar Signals to understand our own software. I want to get into the rhythm and do more of these.
What do you think? How would you have troubleshooted this, and could you have understood what's going on in less than 1min?
What do you think? How would you have troubleshooted this, and could you have understood what's going on in less than 1min?
Reposted by Frederic Branczyk
Good morning #KubeCon!
We are excited to meet all of you and talk about performance! 🚀
Stop by our booth at 1360 next to the Demo Theatre to learn more about CPU, GPU, and memory profiling in production! 🧑💻
We are excited to meet all of you and talk about performance! 🚀
Stop by our booth at 1360 next to the Demo Theatre to learn more about CPU, GPU, and memory profiling in production! 🧑💻
November 11, 2025 at 5:55 PM
Good morning #KubeCon!
We are excited to meet all of you and talk about performance! 🚀
Stop by our booth at 1360 next to the Demo Theatre to learn more about CPU, GPU, and memory profiling in production! 🧑💻
We are excited to meet all of you and talk about performance! 🚀
Stop by our booth at 1360 next to the Demo Theatre to learn more about CPU, GPU, and memory profiling in production! 🧑💻
Is there some global load-balancing product out there that's less expensive than Google Cloud's load balancer? ($0.008/GiB)
November 7, 2025 at 2:40 PM
Is there some global load-balancing product out there that's less expensive than Google Cloud's load balancer? ($0.008/GiB)
You thought docker is causing a lot of wasted space on your disk? Try working on a moderately sized Rust codebase that produces 10 binaries 🫠
November 6, 2025 at 12:22 PM
You thought docker is causing a lot of wasted space on your disk? Try working on a moderately sized Rust codebase that produces 10 binaries 🫠
Never thought I'd say this, but for the first time in a decade, I'm working on a database that can just be queried with SQL, and it's kind of nice 😅
November 6, 2025 at 11:44 AM
Never thought I'd say this, but for the first time in a decade, I'm working on a database that can just be queried with SQL, and it's kind of nice 😅
[email protected] has really outdone himself with the animated parts of this blog post!
Ever wondered how your browser devtools jumps from bundle.min.js:1:27698 → src/index.ts:73:16? 🧊
We unpack the internals of JS Source Maps: Base64 VLQ encoding, mapping deltas, and how your original TS code gets reconstructed from minified bundles.
🔗 www.polarsignals.com/blog/posts/2...
We unpack the internals of JS Source Maps: Base64 VLQ encoding, mapping deltas, and how your original TS code gets reconstructed from minified bundles.
🔗 www.polarsignals.com/blog/posts/2...
The Inner Workings of JavaScript Source Maps
A deep dive into how JavaScript source maps work under the hood, with examples showing how all the pieces fit together.
www.polarsignals.com
November 4, 2025 at 3:11 PM
[email protected] has really outdone himself with the animated parts of this blog post!
I was disappointed with the Unifi lineup from Ubiquiti for some time, but the new products are looking pretty awesome! (APs, 10GbE fiber gateway, 10GbE switches with PoE++)
Mikrotik is lagging behind for the first time in some time for me!
Mikrotik is lagging behind for the first time in some time for me!
October 31, 2025 at 9:30 PM
I was disappointed with the Unifi lineup from Ubiquiti for some time, but the new products are looking pretty awesome! (APs, 10GbE fiber gateway, 10GbE switches with PoE++)
Mikrotik is lagging behind for the first time in some time for me!
Mikrotik is lagging behind for the first time in some time for me!
1% across the board and significantly more for some workloads! Mad props to Canonical for doing the non flashy but impactful important work that only few companies on the planet can!
discourse.ubuntu.com/t/introducin...
discourse.ubuntu.com/t/introducin...
Introducing architecture variants: amd64v3 now available in Ubuntu 25.10
Ubuntu prides itself on being among the most compatible Linux distributions. Compatibility is often a conscious trade-off against bleeding-edge performance. In Ubuntu 25.10, we have added support for…
discourse.ubuntu.com
October 31, 2025 at 8:48 PM
1% across the board and significantly more for some workloads! Mad props to Canonical for doing the non flashy but impactful important work that only few companies on the planet can!
discourse.ubuntu.com/t/introducin...
discourse.ubuntu.com/t/introducin...
We made a bit of a dumb thing early on at Polar Signals where we accidentally created a critical GCS bucket as multi-region. But today I learned about Bucket relocation (launched in 2024). Very timely! docs.cloud.google.com/storage/docs...
Bucket relocation | Cloud Storage | Google Cloud Documentation
This document describes the Cloud Storage bucket relocation service that relocates buckets serverlessly between geographic locations. Using bucket relocation, you can move an existing bucket from one…
docs.cloud.google.com
October 31, 2025 at 2:52 PM
We made a bit of a dumb thing early on at Polar Signals where we accidentally created a critical GCS bucket as multi-region. But today I learned about Bucket relocation (launched in 2024). Very timely! docs.cloud.google.com/storage/docs...
I noticed that cross-zone traffic has once again crept up on us at Polar Signals, so I dusted off kubezonnet, and implemented a few new features/bug fixes as I was troubleshooting.
I present v0.2.0, which we just used to cut 62% of cross-zone traffic. Real money saved.
github.com/polarsignals...
I present v0.2.0, which we just used to cut 62% of cross-zone traffic. Real money saved.
github.com/polarsignals...
Release v0.2.0 · polarsignals/kubezonnet
What's Changed Fix bug about packet sizes being read incorrectly, causing especially small sizes to be inflated. f79dc1b Add port information to flow logs. c136e7c New Contributors @metalmatze m...
github.com
October 31, 2025 at 12:20 PM
I noticed that cross-zone traffic has once again crept up on us at Polar Signals, so I dusted off kubezonnet, and implemented a few new features/bug fixes as I was troubleshooting.
I present v0.2.0, which we just used to cut 62% of cross-zone traffic. Real money saved.
github.com/polarsignals...
I present v0.2.0, which we just used to cut 62% of cross-zone traffic. Real money saved.
github.com/polarsignals...
We're on the HN front page with our post on the NVIDIA GPU profiler we created. Please help us spread the word and comment if you have any questions!
October 29, 2025 at 1:11 PM
We're on the HN front page with our post on the NVIDIA GPU profiler we created. Please help us spread the word and comment if you have any questions!
It's absolutely crazy what computers are capable of when we produce software that works with hardware in mind.
October 29, 2025 at 12:42 PM
It's absolutely crazy what computers are capable of when we produce software that works with hardware in mind.
The NVIDIA RTX Pro 4000 Blackwell with 24GB is looking quite interesting as a workstation GPU as an alternative to a Spark DGX. Obviously not quite apples to apples with different memory speed and sizes but still, 24gb is large enough to do interesting things with.
October 29, 2025 at 7:00 AM
The NVIDIA RTX Pro 4000 Blackwell with 24GB is looking quite interesting as a workstation GPU as an alternative to a Spark DGX. Obviously not quite apples to apples with different memory speed and sizes but still, 24gb is large enough to do interesting things with.
I had a great time talking about GPU profiling at Ubuntu Summit! Thanks everyone for joining!
October 24, 2025 at 10:58 AM
I had a great time talking about GPU profiling at Ubuntu Summit! Thanks everyone for joining!
Reposted by Frederic Branczyk
"We're seeing more Rust recently", "Although 95 % of CUDA workflows use Python"
@brancz.com from @polarsignals.com at #UbuntuSummit2510
#UbuntuSummit
@brancz.com from @polarsignals.com at #UbuntuSummit2510
#UbuntuSummit
October 24, 2025 at 9:12 AM
"We're seeing more Rust recently", "Although 95 % of CUDA workflows use Python"
@brancz.com from @polarsignals.com at #UbuntuSummit2510
#UbuntuSummit
@brancz.com from @polarsignals.com at #UbuntuSummit2510
#UbuntuSummit
If you're at #UbuntuSummit Thursday and Friday, come hear me speak about how we created the world's first open-source low-overhead always-on #NVIDIA #CUDA Profiler! discourse.ubuntu.com/t/infrastruc...
#performance #GPU
#performance #GPU
October 22, 2025 at 2:31 PM
If you're at #UbuntuSummit Thursday and Friday, come hear me speak about how we created the world's first open-source low-overhead always-on #NVIDIA #CUDA Profiler! discourse.ubuntu.com/t/infrastruc...
#performance #GPU
#performance #GPU
I could not be more excited about this. To our knowledge, the world's first open-source low-overhead always-on GPU profiler!
As always, zero-instrumentation and powered by #eBPF!
As always, zero-instrumentation and powered by #eBPF!
🚀 We're super excited to launch NVIDIA CUDA Profiling today! Check out the blog on how we made it happen! 🧊
www.polarsignals.com/blog/posts/2...
www.polarsignals.com/blog/posts/2...
Continuous NVIDIA CUDA Profiling In Production
Low-Overhead Profiling of GPUs with USDT Probes and eBPF
www.polarsignals.com
October 22, 2025 at 2:07 PM
I could not be more excited about this. To our knowledge, the world's first open-source low-overhead always-on GPU profiler!
As always, zero-instrumentation and powered by #eBPF!
As always, zero-instrumentation and powered by #eBPF!
Quick one-hour weekend experiment, and I'm already visualizing our CI infrastructure at Polar Signals as distributed traces in Jaeger! Already gave me insight that I didn't know before (our Rust tests are the critical path, not release binary build times)!
October 18, 2025 at 12:29 PM
Quick one-hour weekend experiment, and I'm already visualizing our CI infrastructure at Polar Signals as distributed traces in Jaeger! Already gave me insight that I didn't know before (our Rust tests are the critical path, not release binary build times)!
It's been that kind of a day ...
October 13, 2025 at 6:48 PM
It's been that kind of a day ...
I don't know why I have to rediscover this over and over, but reducing CI time is such an incredible productivity booster; every second shaved off is worth the investment!
October 13, 2025 at 6:26 PM
I don't know why I have to rediscover this over and over, but reducing CI time is such an incredible productivity booster; every second shaved off is worth the investment!
When did we go from "computers are not great at making use of all those cores" to "a lot of software scales linearly with the number of cores you give it".
Not complaining, just feels like yesterday that we said the former statement a lot.
Not complaining, just feels like yesterday that we said the former statement a lot.
October 12, 2025 at 2:50 PM
When did we go from "computers are not great at making use of all those cores" to "a lot of software scales linearly with the number of cores you give it".
Not complaining, just feels like yesterday that we said the former statement a lot.
Not complaining, just feels like yesterday that we said the former statement a lot.
I have several services that use as much network bandwidth as is available to them on Kubernetes. But I need to prioritize or at least guarantee some amount of bandwidth to a certain component. What are my options other than separate node pools?
October 11, 2025 at 12:38 PM
I have several services that use as much network bandwidth as is available to them on Kubernetes. But I need to prioritize or at least guarantee some amount of bandwidth to a certain component. What are my options other than separate node pools?