-
1
![Snipaste_2024-08-31_16-15-46](https://github.com/user-attachments/assets/5442ca84-26ab-4dd6-b376-8acbef26b81b)
I external/local_xla/xla/stream_executor/cuda/cuda_executor.cc:901] successful NUMA…
-
### Windows Version
Microsoft Windows [Version 10.0.22621.1848]
### WSL Version
1.2.5.0
### Are you using WSL 1 or WSL 2?
- [X] WSL 2
- [ ] WSL 1
### Kernel Version
5.15.90.1
### Distro Versio…
-
**Is your feature request related to a problem? Please describe.**
Some workloads need numa awareness capabilities. On k8s this is covered by: https://kubernetes.io/docs/tasks/administer-cluster/topo…
-
### What is the issue?
My setup is a 4x A100 80GB, 2TB ram, dual intel cpu. Ubuntu server 22.04.
On a previous version of ollama, the model llama3.1:405b was loaded in a reasonable amount of second…
-
Linux kernel counts the following metrics:
- `numa_hint_faults` Records how many NUMA hinting faults were trapped.
- `numa_hint_faults_local` Shows how many of the hinting faults were to local nod…
artem updated
3 weeks ago
-
**Description**
Error `undefined method [] for nil:NilClass` when running `onehost show X` after replacing the host hardware.
Presumably the new hardware has different NUMA capabilities but same …
-
I just measured my machine (Ryzen Milan) and got very similar results as https://github.com/nviennot/core-to-core-latency#dual-amd-epyc-7r13-48-cores-milan-3rd-gen-2021-q1.
However, I was not happy…
-
### Your current environment
The output of `python env.py`
```text
Collecting environment information...
Byte Order: Little Endian
CPU(s): …
-
I actually wrote solution for such situations running exe for those who want threads to be created on different nodes and so that only one CPU is not loaded.
https://github.com/GermanAizek/NUMAye…
-
Currently rayon spawns as many threads as there are *logical* PUs on the machine and does nothing more as far as I can tell. This is subpar. Instead it should look at the hardware topology and make an…