Triton naming

docs/apis/README.md

@@ -821,15 +821,15 @@ TensorflowSpec
 </tr>
 <tr>
 <td>
-<code>tensorrt</code></br>
+<code>triton</code></br>
 <em>
-<a href="#serving.kubeflow.org/v1alpha2.TensorRTSpec">
-TensorRTSpec
+<a href="#serving.kubeflow.org/v1alpha2.TritonSpec">
+TritonSpec
 </a>
 </em>
 </td>
 <td>
-<p>Spec for TensorRT Inference Server (<a href="https://github.com/NVIDIA/tensorrt-inference-server">https://github.com/NVIDIA/tensorrt-inference-server</a>)</p>
+<p>Spec for Triton Inference Server (<a href="https://github.com/NVIDIA/triton-inference-server">https://github.com/NVIDIA/triton-inference-server</a>)</p>
 </td>
 </tr>
 <tr>
@@ -927,7 +927,7 @@ PredictorConfig
 </tr>
 <tr>
 <td>
-<code>tensorrt</code></br>
+<code>triton</code></br>
 <em>
 <a href="#serving.kubeflow.org/v1alpha2.PredictorConfig">
 PredictorConfig
@@ -1153,14 +1153,14 @@ int
 </tr>
 </tbody>
 </table>
-<h3 id="serving.kubeflow.org/v1alpha2.TensorRTSpec">TensorRTSpec
+<h3 id="serving.kubeflow.org/v1alpha2.TritonSpec">TritonSpec
 </h3>
 <p>
 (<em>Appears on:</em>
 <a href="#serving.kubeflow.org/v1alpha2.PredictorSpec">PredictorSpec</a>)
 </p>
 <p>
-<p>TensorRTSpec defines arguments for configuring TensorRT model serving.</p>
+<p>TritonSpec defines arguments for configuring Triton Inference Server.</p>
 </p>
 <table>
 <thead>

docs/predict-api/v2/required_api.md

@@ -7,7 +7,7 @@ By implementing this protocol both
 inference clients and servers will increase their utility and
 portability by being able to operate seamlessly on platforms that have
 standardized around this API. This protocol is endorsed by NVIDIA
-TensorRT Inference Server, TensorFlow Serving, and ONNX Runtime
+Triton Inference Server, TensorFlow Serving, and ONNX Runtime
 Server.
 
 For an inference server to be compliant with this protocol the server

docs/samples/azure/README.md

@@ -4,7 +4,7 @@
 By default, KFServing uses anonymous client to download artifacts. To point to an Azure Blob, specify StorageUri to point to an Azure Blob Storage with the format: 
 ```https://{$STORAGE_ACCOUNT_NAME}.blob.core.windows.net/{$CONTAINER}/{$PATH}```
 
-e.g. https://kfserving.blob.core.windows.net/tensorrt/simple_string/
+e.g. https://kfserving.blob.core.windows.net/triton/simple_string/
 
 ## Using Private Blobs
 KFServing supports authenticating using an Azure Service Principle.

docs/samples/triton/simple_string/README.md

@@ -1,28 +1,28 @@
 
-# Predict on a InferenceService using TensorRT Inference Server
+# Predict on a InferenceService using Triton Inference Server
 ## Setup
 1. Your ~/.kube/config should point to a cluster with [KFServing installed](https://github.com/kubeflow/kfserving/blob/master/docs/DEVELOPER_GUIDE.md#deploy-kfserving).
 2. Your cluster's Istio Ingress gateway must be network accessible.
 
 ## Create the InferenceService
 Apply the CRD
 ```
-kubectl apply -f tensorrt.yaml 
+kubectl apply -f triton.yaml 
 ```
 
 Expected Output
 ```
-inferenceservice.serving.kubeflow.org/tensorrt-simple-string created
+inferenceservice.serving.kubeflow.org/triton-simple-string created
 ```
 
 ## Run a prediction
-Uses the client at: https://docs.nvidia.com/deeplearning/sdk/tensorrt-inference-server-guide/docs/client.html#section-client-api
+Uses the client at: https://docs.nvidia.com/deeplearning/triton-inference-server/user-guide/docs/client_example.html
 
 
 1. setup vars
 
 ```
-SERVICE_HOSTNAME=$(kubectl get ksvc tensorrt-simple-string-predictor-default -o jsonpath='{.status.url}' | cut -d "/" -f 3)
+SERVICE_HOSTNAME=$(kubectl get ksvc triton-simple-string-predictor-default -o jsonpath='{.status.url}' | cut -d "/" -f 3)
 INGRESS_GATEWAY=istio-ingressgateway
 CLUSTER_IP=$(kubectl -n istio-system get service $INGRESS_GATEWAY -o jsonpath='{.status.loadBalancer.ingress[0].ip}')
 echo $CLUSTER_IP
@@ -31,7 +31,7 @@ echo $CLUSTER_IP
 ```
 curl -H "Host: ${SERVICE_HOSTNAME}" http://${CLUSTER_IP}/api/status
 ```
-3. edit /etc/hosts to map the CLUSTER IP to tensorrt-simple-string-predictor-default.default.example.com
+3. edit /etc/hosts to map the CLUSTER IP to triton-simple-string-predictor-default.default.example.com
 4. run the client
 ```
 docker run -e SERVICE_HOSTNAME:$SERVICE_HOSTNAME -it --rm --net=host kcorer/tensorrtserver_client:19.05
@@ -40,7 +40,7 @@ docker run -e SERVICE_HOSTNAME:$SERVICE_HOSTNAME -it --rm --net=host kcorer/tens
 
 You should see output like:
 ```
-root@trantor:/workspace# ./build/simple_string_client -u tensorrt-simple-string.default.example.com
+root@trantor:/workspace# ./build/simple_string_client -u triton-simple-string.default.example.com
 0 + 1 = 1
 0 - 1 = -1
 1 + 1 = 2

docs/samples/triton/simple_string/tensorrt.yaml → docs/samples/triton/simple_string/triton.yaml

@@ -1,9 +1,9 @@
 apiVersion: "serving.kubeflow.org/v1alpha2"
 kind: "InferenceService"
 metadata:
-  name: "tensorrt-simple-string"
+  name: "triton-simple-string"
 spec:
   default:
     predictor:
-      tensorrt:
+      triton:
         storageUri: "gs://kfserving-samples/models/tensorrt"

python/kfserving/README.md

@@ -40,7 +40,7 @@ KFServing supports the following storage providers:
     * By default, it uses `S3_ENDPOINT`, `AWS_ACCESS_KEY_ID`, and `AWS_SECRET_ACCESS_KEY` environment variables for user authentication.
 * Azure Blob Storage with the format: "https://{$STORAGE_ACCOUNT_NAME}.blob.core.windows.net/{$CONTAINER}/{$PATH}"
     * By default, it uses anonymous client to download the artifacts.
-    * For e.g. https://kfserving.blob.core.windows.net/tensorrt/simple_string/
+    * For e.g. https://kfserving.blob.core.windows.net/triton/simple_string/
 * Local filesystem either without any prefix or with a prefix "file://". For example:
     * Absolute path: `/absolute/path` or `file:///absolute/path`
     * Relative path: `relative/path` or `file://relative/path`
@@ -94,7 +94,7 @@ Class | Method |  Description
  - [V1alpha2PyTorchSpec](docs/V1alpha2PyTorchSpec.md)
  - [V1alpha2SKLearnSpec](docs/V1alpha2SKLearnSpec.md)
  - [V1alpha2StatusConfigurationSpec](docs/V1alpha2StatusConfigurationSpec.md)
- - [V1alpha2TensorRTSpec](docs/V1alpha2TensorRTSpec.md)
+ - [V1alpha2TritonSpec](docs/V1alpha2TritonSpec.md)
  - [V1alpha2TensorflowSpec](docs/V1alpha2TensorflowSpec.md)
  - [V1alpha2TransformerSpec](docs/V1alpha2TransformerSpec.md)
  - [V1alpha2XGBoostSpec](docs/V1alpha2XGBoostSpec.md)

python/kfserving/test/test_azure_storage.py

@@ -43,7 +43,7 @@ def get_call_args(call_args_list):
 def test_blob(mock_storage, mock_makedirs): # pylint: disable=unused-argument
 
     # given
-    blob_path = 'https://kfserving.blob.core.windows.net/tensorrt/simple_string/'
+    blob_path = 'https://kfserving.blob.core.windows.net/triton/simple_string/'
     paths = ['simple_string/1/model.graphdef', 'simple_string/config.pbtxt']
     mock_blob = create_mock_blob(mock_storage, paths)
 
@@ -53,8 +53,8 @@ def test_blob(mock_storage, mock_makedirs): # pylint: disable=unused-argument
     # then
     arg_list = get_call_args(mock_blob.get_blob_to_path.call_args_list)
     assert arg_list == [
-        ('tensorrt', 'simple_string/1/model.graphdef', 'dest_path/1/model.graphdef'),
-        ('tensorrt', 'simple_string/config.pbtxt', 'dest_path/config.pbtxt')
+        ('triton', 'simple_string/1/model.graphdef', 'dest_path/1/model.graphdef'),
+        ('triton', 'simple_string/config.pbtxt', 'dest_path/config.pbtxt')
         ]
 
     mock_storage.assert_called_with(account_name="kfserving")
@@ -65,7 +65,7 @@ def test_blob(mock_storage, mock_makedirs): # pylint: disable=unused-argument
 def test_secure_blob(mock_storage, mock_get_token, mock_makedirs): # pylint: disable=unused-argument
 
     # given
-    blob_path = 'https://kfsecured.blob.core.windows.net/tensorrt/simple_string/'
+    blob_path = 'https://kfsecured.blob.core.windows.net/triton/simple_string/'
     mock_blob = mock_storage.return_value
     mock_blob.list_blobs.side_effect = AzureMissingResourceHttpError("fail auth", 404)
     mock_get_token.return_value = "some_token"