mirror of
https://github.com/hanzo-ml/trainer.git
synced 2026-09-24 00:31:34 +00:00
Tariq HasanandGitHub
1f57df4484
* ci(boilerplate): enforce copyright headers on YAML files Signed-off-by: tariq-hasan <[email protected]> * ci(boilerplate): enforce year-less header for new files Signed-off-by: tariq-hasan <[email protected]> --------- Signed-off-by: tariq-hasan <[email protected]>
44 lines
1.8 KiB
YAML
44 lines
1.8 KiB
YAML
# Copyright The Kubeflow Authors.
|
|
#
|
|
# Licensed under the Apache License, Version 2.0 (the "License");
|
|
# you may not use this file except in compliance with the License.
|
|
# You may obtain a copy of the License at
|
|
#
|
|
# http://www.apache.org/licenses/LICENSE-2.0
|
|
#
|
|
# Unless required by applicable law or agreed to in writing, software
|
|
# distributed under the License is distributed on an "AS IS" BASIS,
|
|
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
# See the License for the specific language governing permissions and
|
|
# limitations under the License.
|
|
|
|
# This example deploys the LAMMPS Molecular Dynamic Simulator
|
|
# with MPI orchestrated by the Flux workload manager on 4 nodes.
|
|
# The problem size is defined by the coordinates x,y,z, and the
|
|
# parameter file reaxc.hns.
|
|
# The image has the application, LAMMPS, installed (no Flux)
|
|
# A Flux view will be added on the fly by the Kubeflow trainer
|
|
# Each pod ideally maps 1:1 to nodes, encompassing a cluster
|
|
# The underlying abstraction is a JobSet with a headless service
|
|
# Flux supports low-latency with Infiniband, EFA, etc., however
|
|
# standard ethernet is used here.
|
|
apiVersion: trainer.kubeflow.org/v1alpha1
|
|
kind: TrainJob
|
|
metadata:
|
|
name: lammps-flux
|
|
spec:
|
|
# Reference the pre-defined runtime by name
|
|
runtimeRef:
|
|
name: flux-distributed
|
|
trainer:
|
|
numNodes: 2
|
|
numProcPerNode: 4
|
|
image: ghcr.io/converged-computing/metric-lammps:latest
|
|
# You do not need to write "flux run, etc" here. It will be wrapped
|
|
command: ["lmp", "-v", "x", "4", "-v", "y", "4", "-v", "z", "4", "-in", "in.reaxc.hns", "-nocite"]
|
|
# See flux-interactive.yaml for running the above interactively.
|
|
# This is how we match the view (operating system and version) of the initContainer to install Flux
|
|
env:
|
|
- name: FLUX_VIEW_IMAGE
|
|
value: ghcr.io/converged-computing/flux-view-ubuntu:tag-jammy
|