hiro/ray
1
0
Fork 0
mirror of https://github.com/vale981/ray synced 2025-03-05 10:01:43 -05:00
Code Issues Projects Releases Packages Wiki Activity Actions

[clusters][docs] Provide urls to content, fix typos (#27936)

Browse source
This commit is contained in:
Dmitri Gekhtman 2022-08-18 11:33:04 -07:00 • committed by scv119
parent a08ed6ba75
commit 17a8db048f
6 changed files with 35 additions and 63 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

23
doc/source/cluster/kubernetes/examples/ml-example.md
View file

@ -55,10 +55,7 @@ Broadly speaking, it is more efficient to use a few large Ray pods than many sma
We recommend taking a look at the [config file][ConfigLink] applied in the following command. We recommend taking a look at the [config file][ConfigLink] applied in the following command.
```shell ```shell
# Starting from the parent directory of cloned Ray master, kubectl apply -f https://raw.githubusercontent.com/ray-project/ray/releases/2.0.0/doc/source/cluster/kubernetes/configs/xgboost-benchmark.yaml
pushd ray/doc/source/cluster/kubernetes/configs/
kubectl apply -f xgboost-benchmark.yaml
popd
``` ```
A Ray head pod and 9 Ray worker pods will be created. A Ray head pod and 9 Ray worker pods will be created.
@ -66,7 +63,7 @@ A Ray head pod and 9 Ray worker pods will be created.
```{admonition} Optional: Deploying an autoscaling Ray cluster ```{admonition} Optional: Deploying an autoscaling Ray cluster
If you've set up an autoscaling node group or pool, you may wish to deploy If you've set up an autoscaling node group or pool, you may wish to deploy
an autoscaling cluster by applying the config `xgboost-benchmark-autoscaler.yaml`. an autoscaling cluster by applying the config [xgboost-benchmark-autoscaler.yaml][ConfigLinkAutoscaling].
One Ray head pod will be created. Once the workload starts, the Ray autoscaler will trigger One Ray head pod will be created. Once the workload starts, the Ray autoscaler will trigger
creation of Ray worker pods. Kubernetes autoscaling will then create nodes to place the Ray pods. creation of Ray worker pods. Kubernetes autoscaling will then create nodes to place the Ray pods.
``` ```
@ -100,13 +97,13 @@ We'll use the {ref}`Ray Job Python SDK <ray-job-sdk>` to submit the XGBoost work
``` ```
To submit the workload, run the above Python script. To submit the workload, run the above Python script.
The script is available in the Ray repository. The script is available [in the Ray repository][XGBSubmit].
```shell ```shell
# From the parent directory of cloned Ray master. # Download the above script.
pushd ray/doc/source/cluster/doc_code/ curl https://raw.githubusercontent.com/ray-project/ray/releases/2.0.0/doc/source/cluster/doc_code/xgboost_submit.py -o xgboost_submit.py
# Run the script.
python xgboost_submit.py python xgboost_submit.py
popd
``` ```
### Observe progress. ### Observe progress.
@ -118,7 +115,7 @@ Use the following tools to observe its progress.
To follow the job's logs, use the command printed by the above submission script. To follow the job's logs, use the command printed by the above submission script.
```shell ```shell
# Subsitute the Ray Job's submission id. # Substitute the Ray Job's submission id.
ray job logs 'raysubmit_xxxxxxxxxxxxxxxx' --follow ray job logs 'raysubmit_xxxxxxxxxxxxxxxx' --follow
``` ```
@ -184,6 +181,6 @@ kubectl delete raycluster raycluster-xgboost-benchmark
If you're on a public cloud, don't forget to clean up the underlying If you're on a public cloud, don't forget to clean up the underlying
node group and/or Kubernetes cluster. node group and/or Kubernetes cluster.
<!-- TODO: Fix this --> [ConfigLink]:https://raw.githubusercontent.com/ray-project/ray/releases/2.0.0/doc/source/cluster/kubernetes/configs/xgboost-benchmark.yaml
<!-- [ConfigLink]: https://raw.githubusercontent.com/ray-project/ray/291bba69fb90ee5e8401540ef55b7b74dd13f5c5/doc/source/cluster/ray-clusters-on-kubernetes/configs/xgboost-benchmark-autoscaler.yaml --> [ConfigLinkAutoscaling]: https://raw.githubusercontent.com/ray-project/ray/releases/2.0.0/doc/source/cluster/kubernetes/configs/xgboost-benchmark-autoscaler.yaml
[ConfigLink]: https://github.com/ray-project/ray/tree/master/doc/source/cluster/ [XGBSubmit]: https://github.com/ray-project/ray/blob/releases/2.0.0/doc/source/cluster/doc_code/xgboost_submit.py

42
doc/source/cluster/kubernetes/getting-started.ipynb
View file

@ -79,7 +79,7 @@
"(kuberay-operator-deploy)=\n", "(kuberay-operator-deploy)=\n",
"## Deploying the KubeRay operator\n", "## Deploying the KubeRay operator\n",
"\n", "\n",
"Deploy the KubeRay Operator by cloning the KubeRay repo and applying the relevant configuration files from the master branch. " "Deploy the KubeRay Operator by applying the relevant configuration files from the KubeRay GitHub repo. "
] ]
}, },
{ {
@ -89,12 +89,12 @@
"metadata": {}, "metadata": {},
"outputs": [], "outputs": [],
"source": [ "source": [
"! git clone https://github.com/ray-project/kuberay -b release-0.3\n",
"\n",
"# This creates the KubeRay operator and all of the resources it needs.\n", "# This creates the KubeRay operator and all of the resources it needs.\n",
"! kubectl create -k kuberay/ray-operator/config/default\n", "! kubectl create -k \"github.com/ray-project/kuberay/ray-operator/config/default?ref=v0.3.0&timeout=90s\"\n",
"\n", "\n",
"# Note that we must use \"kubectl create\" in the above command. \"kubectl apply\" will not work due to https://github.com/ray-project/kuberay/issues/271" "# Note that we must use \"kubectl create\" in the above command. \"kubectl apply\" will not work due to https://github.com/ray-project/kuberay/issues/271\n",
"\n",
"# You may alternatively clone the KubeRay GitHub repo and deploy the operator's configuration from your local file system."
] ]
}, },
{ {
@ -156,8 +156,8 @@
"metadata": {}, "metadata": {},
"outputs": [], "outputs": [],
"source": [ "source": [
"# Deploy the Ray Cluster CR:\n", "# Deploy a sample Ray Cluster CR from the KubeRay repo:\n",
"! kubectl apply -f kuberay/ray-operator/config/samples/ray-cluster.autoscaler.yaml\n", "! kubectl apply -f https://raw.githubusercontent.com/ray-project/kuberay/release-0.3/ray-operator/config/samples/ray-cluster.autoscaler.yaml\n",
"\n", "\n",
"# This Ray cluster is named `raycluster-autoscaler` because it has optional Ray Autoscaler support enabled." "# This Ray cluster is named `raycluster-autoscaler` because it has optional Ray Autoscaler support enabled."
] ]
@ -306,8 +306,7 @@
"execution_count": 2, "execution_count": 2,
"id": "d3dae5fd", "id": "d3dae5fd",
"metadata": {}, "metadata": {},
"outputs": [ "outputs": [],
],
"source": [ "source": [
"! kubectl get service raycluster-autoscaler-head-svc\n", "! kubectl get service raycluster-autoscaler-head-svc\n",
"\n", "\n",
@ -379,7 +378,7 @@
"\n", "\n",
"### Deleting a Ray Cluster\n", "### Deleting a Ray Cluster\n",
"\n", "\n",
"To delete the Ray Cluster we deployed in this example, you can run either of the following commands." "To delete the Ray Cluster we deployed in this example, run the following command."
] ]
}, },
{ {
@ -393,25 +392,6 @@
"! kubectl delete raycluster raycluster-autoscaler" "! kubectl delete raycluster raycluster-autoscaler"
] ]
}, },
{
"cell_type": "markdown",
"id": "d7aa0221",
"metadata": {},
"source": [
"**OR**"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "112e6d2e",
"metadata": {},
"outputs": [],
"source": [
"# Delete by reference to the yaml file we used to define the RayCluster CR \n",
"! kubectl delete -f kuberay/ray-operator/config/samples/ray-cluster.autoscaler.yaml"
]
},
{ {
"cell_type": "markdown", "cell_type": "markdown",
"id": "0de87d9d", "id": "0de87d9d",
@ -450,7 +430,7 @@
"metadata": {}, "metadata": {},
"outputs": [], "outputs": [],
"source": [ "source": [
"! kubectl delete -k kuberay/ray-operator/config/default" "! kubectl delete -k \"github.com/ray-project/kuberay/ray-operator/config/default?ref=v0.3.0&timeout=90s\""
] ]
}, },
{ {
@ -490,7 +470,7 @@
"name": "python", "name": "python",
"nbconvert_exporter": "python", "nbconvert_exporter": "python",
"pygments_lexer": "ipython3", "pygments_lexer": "ipython3",
"version": "3.7.11" "version": "3.8.10"
} }
}, },
"nbformat": 4, "nbformat": 4,

9
doc/source/cluster/kubernetes/user-guides/configuring-autoscaling.md
View file

@ -44,13 +44,12 @@ First, follow the [quickstart guide](kuberay-quickstart) to create an autoscalin
```bash ```bash
# Optionally use kind to run the examples locally. # Optionally use kind to run the examples locally.
# kind create cluster # $ kind create cluster
$ git clone https://github.com/ray-project/kuberay -b release-0.3
# Create the KubeRay operator. # Create the KubeRay operator.
$ kubectl create -k kuberay/ray-operator/config/default $ kubectl create -k "github.com/ray-project/kuberay/ray-operator/config/default?ref=v0.3.0&timeout=90s"
# Create an autoscaling Ray cluster. # Create an autoscaling Ray cluster.
$ kubectl apply -f kuberay/ray-operator/config/samples/ray-cluster.autoscaler.yaml $ kubectl apply -f https://raw.githubusercontent.com/ray-project/kuberay/release-0.3/ray-operator/config/samples/ray-cluster.autoscaler.yaml
``` ```
Now, we can run a Ray program on the head pod that uses [``request_resources``](ref-autoscaler-sdk) to scale the cluster to a total of 3 CPUs. The head and worker pods in our [example cluster config](https://github.com/ray-project/kuberay/blob/master/ray-operator/config/samples/ray-cluster.autoscaler.yaml) each have a capacity of 1 CPU, and we specified a minimum of 1 worker pod. Thus, the request should trigger upscaling of one additional worker pod. Now, we can run a Ray program on the head pod that uses [``request_resources``](ref-autoscaler-sdk) to scale the cluster to a total of 3 CPUs. The head and worker pods in our [example cluster config](https://github.com/ray-project/kuberay/blob/master/ray-operator/config/samples/ray-cluster.autoscaler.yaml) each have a capacity of 1 CPU, and we specified a minimum of 1 worker pod. Thus, the request should trigger upscaling of one additional worker pod.
@ -66,7 +65,7 @@ $ kubectl get pods --selector=ray.io/cluster=raycluster-autoscaler --selector=ra
Then, we can run the Ray program using ``kubectl exec``: Then, we can run the Ray program using ``kubectl exec``:
```bash ```bash
$ kubectl exec raycluster-autoscaler-head-xxxxx -it -c ray-head -- python -c \"import ray; ray.init(); ray.autoscaler.sdk.request_resources(num_cpus=3) $ kubectl exec raycluster-autoscaler-head-xxxxx -it -c ray-head -- python -c "import ray; ray.init(); ray.autoscaler.sdk.request_resources(num_cpus=3)"
``` ```
The last command should have triggered Ray pod upscaling. To confirm the new worker pod is up, let's query the RayCluster's pods again: The last command should have triggered Ray pod upscaling. To confirm the new worker pod is up, let's query the RayCluster's pods again:

9
doc/source/cluster/kubernetes/user-guides/logging.md
View file

@ -122,10 +122,7 @@ for instructions on this step.
Now, run the following commands to deploy the Fluent Bit ConfigMap and a single-pod RayCluster with Now, run the following commands to deploy the Fluent Bit ConfigMap and a single-pod RayCluster with
a Fluent Bit sidecar. a Fluent Bit sidecar.
```shell ```shell
# Starting from the parent of cloned Ray master. kubectl apply -f https://raw.githubusercontent.com/ray-project/ray/releases/2.0.0/doc/source/cluster/kubernetes/configs/ray-cluster.log.yaml
pushd ray/doc/source/cluster/kubernetes/configs/
kubectl apply -f ray-cluster.log.yaml
popd
``` ```
Determine the Ray pod's name with Determine the Ray pod's name with
@ -145,6 +142,4 @@ kubectl logs raycluster-complete-logs-head-xxxxx -c fluentbit
[Fluentd]: https://docs.fluentd.org/ [Fluentd]: https://docs.fluentd.org/
[Promtail]: https://grafana.com/docs/loki/latest/clients/promtail/ [Promtail]: https://grafana.com/docs/loki/latest/clients/promtail/
[KubDoc]: https://kubernetes.io/docs/concepts/cluster-administration/logging/ [KubDoc]: https://kubernetes.io/docs/concepts/cluster-administration/logging/
<!-- TODO: fix this --> [ConfigLink]: https://raw.githubusercontent.com/ray-project/ray/releases/2.0.0/doc/source/cluster/kubernetes/configs/ray-cluster.log.yaml
[ConfigLink]: https://github.com/ray-project/ray/tree/master/doc/source/cluster/
<!-- [ConfigLink]: https://raw.githubusercontent.com/ray-project/ray/779e9f7c5733ef9a471ad2bb61723158ff942e92/doc/source/cluster/ray-clusters-on-kubernetes/configs/ray-cluster.log.yaml -->

10
doc/source/cluster/vms/examples/ml-example.md
View file

@ -70,13 +70,13 @@ We'll use the {ref}`Ray Job Python SDK <ray-job-sdk>` to submit the XGBoost work
``` ```
To submit the workload, run the above Python script. To submit the workload, run the above Python script.
The script is also available in the Ray repository. The script is available [in the Ray repository][XGBSubmit].
```shell ```shell
# From the parent directory of cloned Ray master. # Download the above script.
pushd ray/doc/source/cluster/doc_code/ curl https://raw.githubusercontent.com/ray-project/ray/releases/2.0.0/doc/source/cluster/doc_code/xgboost_submit.py -o xgboost_submit.py
# Run the script.
python xgboost_submit.py python xgboost_submit.py
popd
``` ```
### Observe progress ### Observe progress
@ -130,3 +130,5 @@ Delete your Ray cluster with the following command:
```shell ```shell
ray down -y cluster.yaml ray down -y cluster.yaml
``` ```
[XGBSubmit]: https://github.com/ray-project/ray/blob/releases/2.0.0/doc/source/cluster/doc_code/xgboost_submit.py

5
python/ray/autoscaler/_private/autoscaler.py
View file

@ -151,9 +151,8 @@ class StandardAutoscaler:
`ray start --head --autoscaling-config=/path/to/config.yaml` on a instance `ray start --head --autoscaling-config=/path/to/config.yaml` on a instance
that has permission to launch other instances, or you can also use `ray up that has permission to launch other instances, or you can also use `ray up
/path/to/config.yaml` from your laptop, which will configure the right /path/to/config.yaml` from your laptop, which will configure the right
AWS/Cloud roles automatically. See the documentation for a full definition AWS/Cloud roles automatically. See the Ray documentation
of autoscaling behavior: (https://docs.ray.io/en/latest/) for a full definition of autoscaling behavior.
https://docs.ray.io/en/master/cluster/autoscaling.html
StandardAutoscaler's `update` method is periodically called in StandardAutoscaler's `update` method is periodically called in
`monitor.py`'s monitoring loop. `monitor.py`'s monitoring loop.