one could say the same about genocides, but clearly all the museums, exhibits aren't solving the root cause Β―\_(γ)_/Β―
07.08.2025 17:55 β π 0 π 0 π¬ 1 π 0@ahmet.dev.bsky.social
working on compute infra at LinkedIn. ex-twitter/googlecloud/azure dist sys enthusiast github.com/ahmetb
one could say the same about genocides, but clearly all the museums, exhibits aren't solving the root cause Β―\_(γ)_/Β―
07.08.2025 17:55 β π 0 π 0 π¬ 1 π 0Anthropic is leading something?
07.08.2025 17:52 β π 0 π 0 π¬ 1 π 0I absolutely love the table of contents design and animations you have.
07.08.2025 16:46 β π 1 π 0 π¬ 0 π 0I cooked yesterday's block party potluck recipe with LLM, does it count?
06.08.2025 14:50 β π 1 π 0 π¬ 0 π 0Admiralty's been out there forever and its proxy/chaperon idea is sadly one of the most nonscalable approaches.
06.08.2025 14:49 β π 0 π 0 π¬ 1 π 0Anyone have recommended readings on any scalable k8s multi-cluster work API implementations? Planning to roll out a similar model on our own and I want to study the state of the art and the rationale behind the existing model.
05.08.2025 17:55 β π 2 π 0 π¬ 2 π 0You gotta read it in his voice too
05.08.2025 03:28 β π 1 π 0 π¬ 1 π 0Shoulda asked about that blue dye shit
05.08.2025 03:25 β π 1 π 0 π¬ 1 π 0#KubeCon schedule is out! Team LinkedIn will give 3 talks this year (myself included):
1οΈβ£ Making Application Rollouts Observable (lnkd.in/gmnM-8Z4)
2οΈβ£ How We Upgrade the Whole ML Training Fleet Bi-weekly(lnkd.in/gG_a_Etc)
3οΈβ£ Evicted! All the Ways Kubernetes Kills Your Pods
(lnkd.in/ggm9HKZF)
yeah we do that but it would be so much better if we didn't need to. all our microservices and our entire service infra except for CNCF components use push model, and we have to run a otel collector just for those.
30.07.2025 16:43 β π 0 π 0 π¬ 1 π 0Our team at LinkedIn will give 3 talks at the #KubeCon 2025! Looks like we'll be meeting in Atlanta again. ππΌ Stay tuned.
30.07.2025 03:36 β π 10 π 0 π¬ 0 π 0I think this largely assumes OLTP pushing to Prometheus backend, whereas it would be good to have no Prometheus at all in the picture. Admittedly, nearly the whole CNCF ecosystem makes a big assumption around Prometheus, whereas projects like Kyverno catching up to using OTel SDK.
30.07.2025 03:33 β π 2 π 0 π¬ 1 π 0Yeah but how do we change all these off the shelf components to use OTel SDK instead of Prometheus?
30.07.2025 03:13 β π 1 π 0 π¬ 2 π 0Does anyone know why all cloud-native open source ecosystem uses pull model to expose their /metrics (instead of pushing the metrics)? Pull model is really poor when the metric cardinality exposed on a single endpoint increases.
29.07.2025 20:41 β π 2 π 0 π¬ 8 π 1pedantic point: creation idea of Israel and mass Jewish migration to the British Mandate Palestine predate WW2 by many decades (Theodor Herzl, First Zionist Congress, the Uganda plan, Balfour declaration, ...)
25.07.2025 00:44 β π 0 π 0 π¬ 1 π 0Weβve been systematically evaluating the scalability of our fleetβs control planes, and the recent work on storage layer scalability in the Kubernetes community is going to prove pivotal as we scale further!
Check out what weβve been upto for the past couple of weeks β₯οΈ
As we test Kubernetes clusters beyond the supported 5,000+ nodes limit at LinkedIn, we're finding out 'List API' calls are a bottleneck in cluster reliability. So I'm publishing a deep dive about why this is a problem and what you can do about them. ahmet.im/blog/kuberne...
09.07.2025 20:38 β π 24 π 7 π¬ 0 π 1no man stop right there. it's a timesink.
07.07.2025 04:39 β π 1 π 0 π¬ 1 π 0Yeah, we also plumb all the data to mysql over time and build UIs on top. This is more for ad hoc surgical queries for nodes matching a particular label (plus I wanna be able to use this in my next jobs too π)
03.07.2025 00:50 β π 2 π 0 π¬ 0 π 0have you considered Hugo instead? it's not extremely feature rich or JS-heavy like Docusaurus, and is a bit more barebones, yet works well in sites like k8s.io.
03.07.2025 00:40 β π 2 π 0 π¬ 0 π 0Open sourcing a new kubectl plugin called `kubectl node-resource` that queries allocations/utilizations for all/subset of the Kubernetes nodes. It has optimizations for large clusters. We've needed this in our set up to quickly answer questions. github.com/ahmetb/kubec...
23.06.2025 21:43 β π 43 π 12 π¬ 1 π 0Madhav and Ahmet in the office
Glad to have @madhavjivrajani.bsky.social join us at LinkedIn compute team this summer. Iβm constantly impressed by the depth of his kube-apiserver expertise.
23.06.2025 16:36 β π 12 π 1 π¬ 2 π 0Do you have any strava or mapmyride routes of this? Would love to give it a shot.
31.05.2025 20:49 β π 2 π 0 π¬ 1 π 0no :D
30.05.2025 16:30 β π 0 π 0 π¬ 0 π 0DAYS SINCE ETCD BACKUPS SAVED OUR ASS IN PRODUCTION: 0
If you manage your Kubernetes clusters and think you wonβt lose your etcd, youβre almost certainly wrong. It takes years for the disaster to happen and youβll be glad you have invested in building a backup/restore system.
is it officially reversed? I couldn't find much in writing about it.
24.05.2025 04:20 β π 0 π 0 π¬ 1 π 0@kelseyhightower.com hi Kelsey, what was your toto model again? canβt find after the twitter nuke βΉοΈ
11.05.2025 17:42 β π 2 π 0 π¬ 1 π 0what does it even do lol? do we turn off the street lights April 21-28? I feel like a whereas is missing in there.
29.04.2025 02:03 β π 2 π 0 π¬ 0 π 0is swarm beta coming anytime soon?
16.04.2025 20:32 β π 1 π 0 π¬ 1 π 0