Frederic Branczyk
@brancz.com
Founder polarsignals.com 🧊 ❄️ Building parca.dev, Prometheus maintainer. Distributed systems, databases, and performance engineering. Actually born in Berlin. (he/him)
#FOSDEM was a ton of fun this year! Check out our recap if you missed any of our talks!
(also FOSDEM is always a better time if you don't catch fosdem-flu which thankfully I didn't this year!)
(also FOSDEM is always a better time if you don't catch fosdem-flu which thankfully I didn't this year!)
ICYMI We have a recap of what we talked about at #FOSDEM! 🧇 🧊
If you're interested in databases, eBPF or Rust this is worth checking out!
www.polarsignals.com/blog/posts/2...
If you're interested in databases, eBPF or Rust this is worth checking out!
www.polarsignals.com/blog/posts/2...
ICYMI: Polar Signals at FOSDEM
Recap of the talks we gave at FOSDEM developer conference
www.polarsignals.com
February 10, 2026 at 3:12 PM
#FOSDEM was a ton of fun this year! Check out our recap if you missed any of our talks!
(also FOSDEM is always a better time if you don't catch fosdem-flu which thankfully I didn't this year!)
(also FOSDEM is always a better time if you don't catch fosdem-flu which thankfully I didn't this year!)
Do I know anyone who has successfully set up Google Workspace user sync and SSO with Azure (as in Google Workspace is the primary)? I could use some help (support has not been helpful).
February 10, 2026 at 10:59 AM
Do I know anyone who has successfully set up Google Workspace user sync and SSO with Azure (as in Google Workspace is the primary)? I could use some help (support has not been helpful).
Reposted by Frederic Branczyk
Polar Signals at #FOSDEM 2026 day 2!
Both talks in the Rust room.
Profiling Rust applications with Parca
Brennan at 12:30 @ UB2.252A (Rust)
fosdem.org/2026/schedul...
Frederic at 16:00 @ UB2.252A (Rust)
Random seeds and state machines: An approach to DST in Rust
fosdem.org/2026/schedul...
Both talks in the Rust room.
Profiling Rust applications with Parca
Brennan at 12:30 @ UB2.252A (Rust)
fosdem.org/2026/schedul...
Frederic at 16:00 @ UB2.252A (Rust)
Random seeds and state machines: An approach to DST in Rust
fosdem.org/2026/schedul...
February 1, 2026 at 9:00 AM
Polar Signals at #FOSDEM 2026 day 2!
Both talks in the Rust room.
Profiling Rust applications with Parca
Brennan at 12:30 @ UB2.252A (Rust)
fosdem.org/2026/schedul...
Frederic at 16:00 @ UB2.252A (Rust)
Random seeds and state machines: An approach to DST in Rust
fosdem.org/2026/schedul...
Both talks in the Rust room.
Profiling Rust applications with Parca
Brennan at 12:30 @ UB2.252A (Rust)
fosdem.org/2026/schedul...
Frederic at 16:00 @ UB2.252A (Rust)
Random seeds and state machines: An approach to DST in Rust
fosdem.org/2026/schedul...
@thorfour.dev telling #FOSDEM about our new rust-based database (with lots of open source inside!)
January 31, 2026 at 4:46 PM
@thorfour.dev telling #FOSDEM about our new rust-based database (with lots of open source inside!)
@gnurizen.bsky.social showing us how we can get a memory profile just when your process is about to be killed by the OOM killer.
January 31, 2026 at 11:35 AM
@gnurizen.bsky.social showing us how we can get a memory profile just when your process is about to be killed by the OOM killer.
All of this is right up my alley! Would love to meet!
FWIW the Polar Signals crew will be around the eBPF, Databases, and Rust devroom pretty much throughout.
FWIW the Polar Signals crew will be around the eBPF, Databases, and Rust devroom pretty much throughout.
January 30, 2026 at 3:30 PM
All of this is right up my alley! Would love to meet!
FWIW the Polar Signals crew will be around the eBPF, Databases, and Rust devroom pretty much throughout.
FWIW the Polar Signals crew will be around the eBPF, Databases, and Rust devroom pretty much throughout.
Very excited for this year! Who else will be there?
January 27, 2026 at 5:14 PM
Very excited for this year! Who else will be there?
I (jumping in for @asubiotto.com) will talk about our approach to Deterministic Simulation Testing (DST) in Rust. Or how we ensure that the most critical bugs never make it to production.
fosdem.org/2026/schedul...
fosdem.org/2026/schedul...
FOSDEM 2026 - Random seeds and state machines: An approach to deterministic simulation testing in Rust
fosdem.org
January 27, 2026 at 5:14 PM
I (jumping in for @asubiotto.com) will talk about our approach to Deterministic Simulation Testing (DST) in Rust. Or how we ensure that the most critical bugs never make it to production.
fosdem.org/2026/schedul...
fosdem.org/2026/schedul...
Brennan will talk about profiling Rust with Parca, focusing on all the awesome extra things possible with Rust!
fosdem.org/2026/schedul...
fosdem.org/2026/schedul...
FOSDEM 2026 - Profiling Rust applications with Parca
fosdem.org
January 27, 2026 at 5:14 PM
Brennan will talk about profiling Rust with Parca, focusing on all the awesome extra things possible with Rust!
fosdem.org/2026/schedul...
fosdem.org/2026/schedul...
@gnurizen.bsky.social will talk about OOMProf, or how we obtain a heap profile just before an OOM occurs.
fosdem.org/2026/schedul...
fosdem.org/2026/schedul...
FOSDEM 2026 - OOMProf: profiling Go heap memory at OOM time
fosdem.org
January 27, 2026 at 5:14 PM
@gnurizen.bsky.social will talk about OOMProf, or how we obtain a heap profile just before an OOM occurs.
fosdem.org/2026/schedul...
fosdem.org/2026/schedul...
@thorfour.dev will talk about our new Diskless database written in Rust with Arrow, DataFusion, and Vortex, and the evolution that got us here.
fosdem.org/2026/schedul...
fosdem.org/2026/schedul...
FOSDEM 2026 - From Disks to Distributed: Our Journey of Database Evolution in the Cloud
fosdem.org
January 27, 2026 at 5:14 PM
@thorfour.dev will talk about our new Diskless database written in Rust with Arrow, DataFusion, and Vortex, and the evolution that got us here.
fosdem.org/2026/schedul...
fosdem.org/2026/schedul...
Reposted by Frederic Branczyk
Symbolization issues don’t have to be mysterious.
Polar Signals Cloud now shows exactly what’s happening with your debuginfo, right from the flamegraph frame: www.polarsignals.com/blog/posts/2...
Polar Signals Cloud now shows exactly what’s happening with your debuginfo, right from the flamegraph frame: www.polarsignals.com/blog/posts/2...
How to check your debuginfo status in Polar Signals Cloud
A faster and easier way to troubleshoot debuginfo processing in Polar Signals Cloud.
www.polarsignals.com
January 22, 2026 at 2:19 PM
Symbolization issues don’t have to be mysterious.
Polar Signals Cloud now shows exactly what’s happening with your debuginfo, right from the flamegraph frame: www.polarsignals.com/blog/posts/2...
Polar Signals Cloud now shows exactly what’s happening with your debuginfo, right from the flamegraph frame: www.polarsignals.com/blog/posts/2...
We've been working on something exciting in the Arrow/DataFusion ecosystem, which finally shipped with yesterday's release of DataFusion. You can now use Run-End-Encoded arrays in group by clauses!
github.com/apache/dataf...
github.com/apache/dataf...
Support Aggregating by `RunArray`s · Issue #16011 · apache/datafusion
Is your feature request related to a problem or challenge? It's currently not possible to aggregate by RunArrays. Example code grouping by a `RunArray` use arrow::array::{Array, Int32Array, RunArra...
github.com
January 13, 2026 at 9:45 AM
We've been working on something exciting in the Arrow/DataFusion ecosystem, which finally shipped with yesterday's release of DataFusion. You can now use Run-End-Encoded arrays in group by clauses!
github.com/apache/dataf...
github.com/apache/dataf...
Reposted by Frederic Branczyk
My EuroRust talk has been uploaded to YT! I cover how we architected our Rust database at @polarsignals.com as a set of state machines to enable deterministic simulation testing and catch hard-to-reproduce bugs before they hit prod.
A new talk is out! 🙌 @asubiotto.com shows how deterministic simulation testing (DST) uses random seeds and state machines to explore countless execution paths—and reliably replay failures when things go wrong 🦀🎲
👉 youtu.be/V12hHoagv-E
👉 youtu.be/V12hHoagv-E
Random Seeds and State Machines - Alfonso Subiotto | EuroRust 2025
Random Seeds and State Machines: An Approach to Deterministic Simulation Testing - Alfonso Subiotto at EuroRust 2025
Have you spent sleepless nights thinking about that heisenbug in production that…
youtu.be
January 7, 2026 at 11:46 AM
My EuroRust talk has been uploaded to YT! I cover how we architected our Rust database at @polarsignals.com as a set of state machines to enable deterministic simulation testing and catch hard-to-reproduce bugs before they hit prod.
Right, I was saying I agree with that perspective from the user (at least currently).
That said, time will tell I think, it’s possible that databases will blur this as they get even better at storing and querying it, but I’m less sure about that aspect.
That said, time will tell I think, it’s possible that databases will blur this as they get even better at storing and querying it, but I’m less sure about that aspect.
January 5, 2026 at 6:40 PM
Right, I was saying I agree with that perspective from the user (at least currently).
That said, time will tell I think, it’s possible that databases will blur this as they get even better at storing and querying it, but I’m less sure about that aspect.
That said, time will tell I think, it’s possible that databases will blur this as they get even better at storing and querying it, but I’m less sure about that aspect.
I think we’re less than a year away from both being the same for databases. TSDBs like Prometheus had 10 years of optimizing, but all of the optimizations are slowly but surely landing in “general purpose” column stores. I agree with the sampled vs unsampled from a user perspective notion though!
January 5, 2026 at 6:10 PM
I think we’re less than a year away from both being the same for databases. TSDBs like Prometheus had 10 years of optimizing, but all of the optimizations are slowly but surely landing in “general purpose” column stores. I agree with the sampled vs unsampled from a user perspective notion though!
What an awesome year at Polar Signals! Some highlights:
📈 >10x data scale
⚡️ Shipped always-on GPU profiling
🦀 Shipped our new Rust/DataFusion/Vortex DB
Feeling lucky to get to work with such an incredible team & customers like Vercel, Notion, Canonical, Turbopuffer and more!
2026, we're ready!
📈 >10x data scale
⚡️ Shipped always-on GPU profiling
🦀 Shipped our new Rust/DataFusion/Vortex DB
Feeling lucky to get to work with such an incredible team & customers like Vercel, Notion, Canonical, Turbopuffer and more!
2026, we're ready!
December 31, 2025 at 5:11 PM
What an awesome year at Polar Signals! Some highlights:
📈 >10x data scale
⚡️ Shipped always-on GPU profiling
🦀 Shipped our new Rust/DataFusion/Vortex DB
Feeling lucky to get to work with such an incredible team & customers like Vercel, Notion, Canonical, Turbopuffer and more!
2026, we're ready!
📈 >10x data scale
⚡️ Shipped always-on GPU profiling
🦀 Shipped our new Rust/DataFusion/Vortex DB
Feeling lucky to get to work with such an incredible team & customers like Vercel, Notion, Canonical, Turbopuffer and more!
2026, we're ready!
Reposted by Frederic Branczyk
We cut cross-zone traffic for SLO monitoring by 90% using Prometheus subquery optimization.
The trick: pre-aggregate data locally instead of shuffling 55M raw samples to Thanos queriers.
Coming soon to Pyrra! 🚀
www.polarsignals.com/blog/posts/2...
The trick: pre-aggregate data locally instead of shuffling 55M raw samples to Thanos queriers.
Coming soon to Pyrra! 🚀
www.polarsignals.com/blog/posts/2...
Do more with less. | Polar Signals
Polar Signals Cloud is an always-on, zero-instrumentation continuous profiling for CPU, GPU, and Memory that helps improve performance, understand incidents, and lower infrastructure costs.
www.polarsignals.com
December 30, 2025 at 3:02 PM
We cut cross-zone traffic for SLO monitoring by 90% using Prometheus subquery optimization.
The trick: pre-aggregate data locally instead of shuffling 55M raw samples to Thanos queriers.
Coming soon to Pyrra! 🚀
www.polarsignals.com/blog/posts/2...
The trick: pre-aggregate data locally instead of shuffling 55M raw samples to Thanos queriers.
Coming soon to Pyrra! 🚀
www.polarsignals.com/blog/posts/2...
Just as I was thinking Unifi was getting ahead of Mikrotik, they release this absolute beast of a switch. Modern computing hardware is pretty awesome.
mikrotik.com/product/crs8...
mikrotik.com/product/crs8...
CRS812 DDQ | MikroTik
This switch is your next leap forward – a powerful, efficient, and cost-effective way to bring 50G, 200G, and 400G into your rack. Quad-core 2 GHz ARM CPU, dual-redundant power supplies and 4x…
mikrotik.com
December 27, 2025 at 4:51 PM
Just as I was thinking Unifi was getting ahead of Mikrotik, they release this absolute beast of a switch. Modern computing hardware is pretty awesome.
mikrotik.com/product/crs8...
mikrotik.com/product/crs8...
Reposted by Frederic Branczyk
A simple 50-line code change saves us 30% of CPU across our whole infrastructure. This is the magic of system-wide profiling with @polarsignals.com. Easy to spot, easy to fix.
performance[vortex-array]: don't call is_valid to count bytes in varbinview by asubiotto · Pull Request #5814 · vortex-data/vortex
I noticed that the better blocks compressor uses count_referenced_bytes which calls is_valid on each view and results in an expensive scalar_at call. This was 30% of our system-wide CPU usage over ...
github.com
December 23, 2025 at 10:48 AM
A simple 50-line code change saves us 30% of CPU across our whole infrastructure. This is the magic of system-wide profiling with @polarsignals.com. Easy to spot, easy to fix.
It's super easy to get started with profiling your NVIDIA CUDA workload, you don't even need to modify it!
🚀 We recently announced our NVIDIA CUDA always-on profiler. Check out this blog on how to try it out super quickly with your CUDA workload today!🧊
www.polarsignals.com/blog/posts/2...
www.polarsignals.com/blog/posts/2...
Profiling NVIDIA CUDA in Kubernetes
The easiest way to get started Profiling CUDA in Kubernetes
www.polarsignals.com
December 18, 2025 at 4:03 PM
It's super easy to get started with profiling your NVIDIA CUDA workload, you don't even need to modify it!
Reposted by Frederic Branczyk
Happy to share that I've been nominated to the Vortex Technical Steering Committee! It's been fun and productive switching to Vortex from Parquet as our storage format @polarsignals.com and I'm excited to continue contributing to the Vortex project.
vortex.dev/blog/novembe...
vortex.dev/blog/novembe...
November Bulletin | Vortex Blog
Overview of all work happening in Vortex
vortex.dev
December 15, 2025 at 3:03 PM
Happy to share that I've been nominated to the Vortex Technical Steering Committee! It's been fun and productive switching to Vortex from Parquet as our storage format @polarsignals.com and I'm excited to continue contributing to the Vortex project.
vortex.dev/blog/novembe...
vortex.dev/blog/novembe...