News for hackers. Stuff that matters.
Re:Why We’re Switching to gRPC by enitihas on Monday May 27, 2019 @02:25PM (#20024209)
Curiously, what do you find lacking in Kubernetes compared to Borg?
Re:Why We’re Switching to gRPC by mehrdada on Monday May 27, 2019 @02:27PM (#20024218)
Scale, for one thing.

Kubernetes is not a bad system but it’s not designed to run Google.

Re:Why We’re Switching to gRPC by Thaxll on Monday May 27, 2019 @04:04PM (#20024899)
What google system runs more than 5k tasks on a single cluster?
Re:Why We’re Switching to gRPC by the-rc on Monday May 27, 2019 @07:33PM (#20026037)
https://github.com/google/cluster-data/blob/master/README.md...

"ClusterData2011_2 provides data from an 12.5k-machine cell over about a month-long period in May 2011."

Re:Why We’re Switching to gRPC by shereadsthenews on Monday May 27, 2019 @08:32PM (#20026253)
10k replicas in a single cell is the default charge-free quota of every individual engineer at that company. It’s basically zero.
Re:Why We’re Switching to gRPC by dilyevsky on Monday May 27, 2019 @06:39PM (#20025758)
Heh, you have no idea. Lots and lots of systems. And 5k wouldn't even register there...
Re:Why We’re Switching to gRPC by Thaxll on Thursday May 30, 2019 @10:18AM (#20051350)
Can't edit my comment anymore but I meant 5k nodes / 150k pods / tasks.
Re:Why We’re Switching to gRPC by int0x80 on Monday May 27, 2019 @02:59PM (#20024457)
Can you expand on that, please?
Re:Why We’re Switching to gRPC by dilyevsky on Monday May 27, 2019 @06:37PM (#20025745)
I used to work on one of Borg teams at Google and now run Kubernetes platform. Nearly every Kubernetes component (node, master, networking) will melt down at fraction of Borg scale. It's not even close.
Re:Why We’re Switching to gRPC by int0x80 on Tuesday May 28, 2019 @04:26AM (#20028231)
thanks for the answer.
Re:Why We’re Switching to gRPC by q3k on Monday May 27, 2019 @03:22PM (#20024642)
https://kubernetes.io/docs/setup/cluster-large/

More specifically, "No more than 5000 nodes" and "No more than 150000 total pods" is fairly limiting to large (Google-large) clusters.

Re:Why We’re Switching to gRPC by int0x80 on Tuesday May 28, 2019 @04:26AM (#20028232)
thanks for the answer.