documentation updates (#60)

openvinotoolkit · May 26, 2022 · 0b4cd95 · 0b4cd95
1 parent 181dd0e
commit 0b4cd95
Show file tree

Hide file tree

Showing 3 changed files with 11 additions and 5 deletions.
diff --git a/README.md b/README.md
@@ -1,6 +1,7 @@
 # Openshift and Kubernetes operator
 
-The Operator installs and manages [OpenVINO model servers](https://github.com/openvinotoolkit/model_server) in an OpenShift cluster and upstream Kubernetes. It enables inference execution at scale and exposing AI models via gRPC and REST API interfaces.
+The Operator installs and manages [OpenVINO model servers](https://github.com/openvinotoolkit/model_server) in an OpenShift cluster and upstream Kubernetes. It enables inference execution at scale and exposes AI models via gRPC and REST API interfaces.
+The operator is using a [model server helm chart](helm-charts/ovms/) which can be also used directly.
 
 The Operator also integrates with the JupyterHub Spawner in Red Hat OpenShift Data Science and Open Data Hub. See detailed instructions below.
 
@@ -20,6 +21,8 @@ The Operator also integrates with the JupyterHub Spawner in Red Hat OpenShift Da
 
 [Notebook parameters](docs/notebook_params.md)
 
+[Helm chart](helm-charts/ovms/README.md)
+
 
 ## Contact
 If you have a question, a feature request, or a bug report, feel free to submit a Github issue.

diff --git a/docs/operator_installation.md b/docs/operator_installation.md
@@ -1,11 +1,13 @@
 # Operator instalation
 
 *Note:* Operator, starting from version 1.0 includes non-compatible changes in the CRD records of `ModelServer` and `Notebook`.
-It is recommended to remove all those custom records before upgrading the operator from v0.2 to v1.0.
+It is recommended to remove all those custom records before upgrading the operator from v0.2 to v1.0. 
+
+*Note:* Source code of the versions prior to 1.0.0 is stored in a [github repo](https://github.com/openvinotoolkit/model_server/tree/v2022.1/extras)
 
 ## Openshift
 
-In the OpenShift [web console](https://docs.openshift.com/container-platform/4.10/web_console/web-console.html) navigate to OperatorHub menu. Search for "OpenVINO™ Toolkit Operator". Then, click the `Install` button.
+In the OpenShift [web console](https://docs.openshift.com/container-platform/4.10/web_console/web-console.html) navigate to the OperatorHub menu. Search for "OpenVINO™ Toolkit Operator". Then, click the `Install` button.
 
 ![installation](install.png)
 
@@ -17,7 +19,7 @@ Find the `OpenVINO Model Server Operator` and click 'Install' button.
 
 ***
 
-## Building and installation from sources
+## Building and installing from sources
 
 Check the [developer guide](developer_guide.md) if you would like to build the operator on your own.
 

diff --git a/helm-charts/ovms/README.md b/helm-charts/ovms/README.md
@@ -102,7 +102,8 @@ That assumes using adequate Kubernetes device plugin from [Intel Device Plugin f
 deployment_parameters:
   resources:
     limits:
-      gpu.intel.com/i915: 1
+      xpu_device: gpu.intel.com/i915
+      xpu_device_quantity: "1"
 ```
 ### Security Context
 OVMS, by default, starts with the security context of `ovms` account which has the pid 5000 and gid 5000. In some cases, it can prevent importing models