
Pierre Zemb
21.3K posts

Pierre Zemb
@PierreZ
🧑💻 Staff Engineer @clever_cloud 🤩 Distributed and Database Systems 🚀 Co-leader @finistdevs 💪 Squash player
Brest, France Katılım Şubat 2009
4K Takip Edilen3.3K Takipçiler
Sabitlenmiş Tweet

✍️ I just published "Best resources to learn about data and distributed systems"
Happy learning 😊
#data #distributedsystem
pierrezemb.fr/posts/distsys-…
English
Pierre Zemb retweetledi
Pierre Zemb retweetledi

"Et si on faisait du simulation-driven development ?" avec Pierre Zemb
Découvrez comment le Deterministic Simulation Testing permet de simuler, rejouer et diagnostiquer les bugs les plus complexes des systèmes distribués. #SunnyTech2026 #SimulationDrivenDevelopment
Français
Pierre Zemb retweetledi
Pierre Zemb retweetledi
Pierre Zemb retweetledi

Our TESS telescope has released its most complete view of the starry sky yet! All those dots? They're confirmed and candidate planets! Since launching in April 2018, TESS has been scanning wide swaths of the sky to build up this mosaic. go.nasa.gov/49MK9hU

English
Pierre Zemb retweetledi
Pierre Zemb retweetledi

This is such a great example of theory vs practice. In theory, UUIDv4 collisions don't happen (generating one million per second, probability of seeing one collision in a year is ~10^-8).
But they have been observed to happen in practice, especially in distributed systems. Why?
v@iavins
UUID v4 collisions are less rare than you think 💣
English
Pierre Zemb retweetledi

I say this all the time. Databases aren’t sold on golf courses anymore.
How you post your measurements and benchmarks matter.
These days a lot more choices are made by engineering teams not executives on the back nine after 6 beers.
The audience is nerds who care.
Sai Srirampur@saisrirampur
@kellabyte Totally makes sense. Telling that story in a real, enlightening and transparent way matters.
English
Pierre Zemb retweetledi

We are back. After one year of quiet building.
Introducing GENE-26.5, our first robotic brain that takes a major step toward human-level capability.
For years, robotics has struggled to learn from the world’s largest and valuable data source: Humans.
Solving it means rethinking the whole stack from the ground up:
- A robotics-native foundation model.
- A 1:1 human-like robotic hand.
- A noninvasive data collection glove for motion, force, and touch.
- A simulator that turns weeks of experiments into minutes.
GENE-26.5 is trained across language, vision, proprioception, tactile, and action. We designed a set of tasks to test how far we can go with this new paradigm.
Fully autonomous, 1x speed, one model, same weights. (Enjoy with sound on)
We are approaching the endgame for robotics.
And this is just a beginning.
English
Pierre Zemb retweetledi
Pierre Zemb retweetledi

I gave a lightning talk at #BugBash 2026, sharing the stage with people whose work shaped how I think about distributed systems. Still processing that.
Wrote up why the correctness decade has started, and why you trust the simulator, not Claude 👀
pierrezemb.fr/posts/bugbash-…
English
Pierre Zemb retweetledi

UPDATE: pgBackRest is likely to receive enough funding be be revived. See github.com/pgbackrest/pgb… for details.
English
Pierre Zemb retweetledi

🚀 Apache Iggy 0.8.0 is out! Wire protocol rewrite, persistent WAL journal, new shard architecture, stronger security, SDK & connectors progress. Clustering keeps accelerating.
iggy.apache.org/blogs/2026/04/…
#iggy #rust #io_uring #vsr #apache #asf #streaming

English
Pierre Zemb retweetledi
Pierre Zemb retweetledi

@kellabyte Don’t be sorry, your work around HoloStore is impressive, and I feel the pain of not having enough hours in the day.
Have fun and take care!
English

@kellabyte Really nice work! Sharding is the reason we wrote an etcd-shim on top of #foundationdb
English

Just fixed throughput/p99 jitter caused from dynamic partition splitting seen in this graph
Even tho HoloStore dynamically splits ranges ala CockroachDB and etcd has one global keyspace HoloStore still has less jitter.
Real early days tho, Holo is a toy, etcd is mature sw.

Kelly Sommers@kellabyte
I've started measuring HoloStore against etcd the consensus datastore that powers Kubernetes I always make sure coordinated omission is correctly accounted for I also benchmark the client. Clients are part of ALL distributed system I also mark HoloStore shard split events
English

New blog post: "The Bombadil Terminal Experiment"
wickstrom.tech/2026-04-30-bom…
English








