chore: Renamed Triton Distributed to Dynamo (#56)

CONTRIBUTING.md

@@ -25,20 +25,20 @@ Contributions intended to add significant new functionality must
 follow a more collaborative path described in the following
 points. Before submitting a large PR that adds a major enhancement or
 extension, be sure to submit a GitHub issue that describes the
-proposed change so that the Triton team can provide feedback.
+proposed change so that the Dynamo team can provide feedback.
 
 - As part of the GitHub issue discussion, a design for your change
   will be agreed upon. An up-front design discussion is required to
   ensure that your enhancement is done in a manner that is consistent
-  with Triton Distributed's overall architecture.
+  with Dynamo's overall architecture.
 
-- The Triton Distributed project is spread across multiple GitHub Repositories.
-  The Triton team will provide guidance about how and where your enhancement
+- The Dynamo project is spread across multiple GitHub Repositories.
+  The Dynamo team will provide guidance about how and where your enhancement
   should be implemented.
 
-- Testing is a critical part of any Triton
+- Testing is a critical part of any Dynamo
   enhancement. You should plan on spending significant time on
-  creating tests for your change. The Triton team will help you to
+  creating tests for your change. The Dynamo team will help you to
   design your testing so that it is compatible with existing testing
   infrastructure.
 
@@ -75,7 +75,7 @@ proposed change so that the Triton team can provide feedback.
 
   - Make sure all tests pass.
 
-- Triton Distributed's default build assumes recent versions of
+- Dynamo's default build assumes recent versions of
   dependencies (CUDA, TensorFlow, PyTorch, TensorRT,
   etc.). Contributions that add compatibility with older versions of
   those dependencies will be considered, but NVIDIA cannot guarantee
@@ -85,7 +85,7 @@ proposed change so that the Triton team can provide feedback.
 - Make sure that you can contribute your work to open source (no
   license and/or patent conflict is introduced by your code).
   You must certify compliance with the
-  [license terms](https://github.com/triton-inference-server/triton-distributed/blob/main/LICENSE)
+  [license terms](https://github.com/ai-dynamo/dynamo/blob/main/LICENSE)
   and sign off on the [Developer Certificate of Origin (DCO)](https://developercertificate.org)
   described below before your pull request (PR) can be merged.
 
@@ -96,7 +96,7 @@ proposed change so that the Triton team can provide feedback.
 
 All pull requests are checked against the
 [pre-commit hooks](https://github.com/pre-commit/pre-commit-hooks)
-located [in the repository's top-level .pre-commit-config.yaml](https://github.com/triton-inference-server/triton-distributed/blob/main/.pre-commit-config.yaml).
+located [in the repository's top-level .pre-commit-config.yaml](https://github.com/ai-dynamo/dynamo/blob/main/.pre-commit-config.yaml).
 The hooks do some sanity checking like linting and formatting.
 These checks must pass to merge a change.
 
@@ -123,7 +123,7 @@ Also you can use vscode extension [GitHub Local Actions](https://marketplace.vis
 
 # Developer Certificate of Origin
 
-Triton Distributed is an open source product released under
+Dynamo is an open source product released under
 the Apache 2.0 license (see either
 [the Apache site](https://www.apache.org/licenses/LICENSE-2.0) or
 the [LICENSE file](./LICENSE)). The Apache 2.0 license allows you
@@ -177,7 +177,7 @@ By making a contribution to this project, I certify that:
     this project or the open source license(s) involved.
 ```
 
-We require that every contribution to Triton Distributed is signed with
+We require that every contribution to Dynamo is signed with
 a Developer Certificate of Origin. Additionally, please use your real name.
 We do not accept anonymous contributors nor those utilizing pseudonyms.
 

container/build.sh

@@ -63,7 +63,7 @@ TENSORRTLLM_BASE_IMAGE_TAG=${TENSORRTLLM_BASE_VERSION}-trtllm-python-py3
 # used in the base image above.
 TENSORRTLLM_BACKEND_REPO_TAG=triton-llm/v0.17.0
 # Set this as 1 to rebuild and replace trtllm backend bits in the container.
-# This will allow building triton distributed container image with custom
+# This will allow building Dynamo container image with custom
 # trt-llm backend repo branch.
 TENSORRTLLM_BACKEND_REBUILD=0
 # Set this as 1 to skip cloning the trt-llm backend repo. If cloning is skipped, trt-llm
@@ -247,7 +247,7 @@ get_options() {
     fi
 
     if [ -z "$TAG" ]; then
-        TAG="--tag triton-distributed:${VERSION}-${FRAMEWORK,,}"
+        TAG="--tag dynamo:${VERSION}-${FRAMEWORK,,}"
         if [ ! -z ${TARGET} ]; then
             TAG="${TAG}-${TARGET}"
         fi
@@ -265,7 +265,7 @@ get_options() {
 
 show_image_options() {
     echo ""
-    echo "Building Triton Distributed Image: '${TAG}'"
+    echo "Building Dynamo Image: '${TAG}'"
     echo ""
     echo "   Base: '${BASE_IMAGE}'"
     echo "   Base_Image_Tag: '${BASE_IMAGE_TAG}'"
@@ -340,7 +340,7 @@ if [ ! -z ${HF_TOKEN} ]; then
     BUILD_ARGS+=" --build-arg HF_TOKEN=${HF_TOKEN} "
 fi
 
-LATEST_TAG="--tag triton-distributed:latest-${FRAMEWORK,,}"
+LATEST_TAG="--tag dynamo:latest-${FRAMEWORK,,}"
 if [ ! -z ${TARGET} ]; then
     LATEST_TAG="${LATEST_TAG}-${TARGET}"
 fi

container/run.sh

@@ -178,7 +178,7 @@ get_options() {
     fi
 
     if [ -z "$IMAGE" ]; then
-        IMAGE="triton-distributed:latest-${FRAMEWORK,,}"
+        IMAGE="dynamo:latest-${FRAMEWORK,,}"
         if [ ! -z ${TARGET} ]; then
             IMAGE="${IMAGE}-${TARGET}"
         fi

examples/python_rs/llm/tensorrt_llm/README.md

@@ -15,9 +15,9 @@ See the License for the specific language governing permissions and
 limitations under the License.
 -->
 
-# TensorRT-LLM Integration with Triton Distributed
+# TensorRT-LLM Integration with Dynamo
 
-This example demonstrates how to use Triton Distributed to serve large language models with the tensorrt_llm engine, enabling efficient model serving with both monolithic and disaggregated deployment options.
+This example demonstrates how to use Dynamo to serve large language models with the tensorrt_llm engine, enabling efficient model serving with both monolithic and disaggregated deployment options.
 
 ## Prerequisites
 
@@ -58,7 +58,7 @@ python3 scripts/build_wheel.py --clean --trt_root /usr/local/tensorrt -a native
 cp build/tensorrt_llm-*.whl /home
 ```
 
-- Build the Triton Distributed container
+- Build the Dynamo container
 ```bash
 # Build image
 ./container/build.sh --base-image gitlab-master.nvidia.com:5005/dl/dgx/tritonserver/tensorrt-llm/amd64 --base-image-tag krish-fix-trtllm-build.23766174
@@ -73,7 +73,7 @@ Alternatively, you can build with latest tensorrt_llm pipeline like below:
 
 ## Launching the Environment
 ```
-# Run image interactively from with the triton distributed root directory.
+# Run image interactively from with the Dynamo root directory.
 ./container/run.sh --framework TENSORRTLLM -it -v /home/:/home/
 
 # Install the TRT-LLM wheel. No need to do this if you are using the latest tensorrt_llm image.
@@ -306,7 +306,7 @@ export ETCD_ENDPOINTS="http://node1:2379,http://node2:2379"
 
 3. Launch the workers from node1 or login node. WORLD_SIZE is similar to single node deployment.
 ```bash
-srun --mpi pmix -N NUM_NODES --ntasks WORLD_SIZE --ntasks-per-node=WORLD_SIZE --no-container-mount-home --overlap --container-image IMAGE --output batch_%x_%j.log --err batch_%x_%j.err --container-mounts PATH_TO_TRITON_DISTRIBUTED:/workspace --container-env=NATS_SERVER,ETCD_ENDPOINTS bash -c 'cd /workspace/examples/python_rs/llm/tensorrt_llm && python3 -m disaggregated.worker --engine_args llm_api_config.yaml -c disaggregated/llmapi_disaggregated_configs/multi_node_config.yaml' &
+srun --mpi pmix -N NUM_NODES --ntasks WORLD_SIZE --ntasks-per-node=WORLD_SIZE --no-container-mount-home --overlap --container-image IMAGE --output batch_%x_%j.log --err batch_%x_%j.err --container-mounts PATH_TO_DYNAMO:/workspace --container-env=NATS_SERVER,ETCD_ENDPOINTS bash -c 'cd /workspace/examples/python_rs/llm/tensorrt_llm && python3 -m disaggregated.worker --engine_args llm_api_config.yaml -c disaggregated/llmapi_disaggregated_configs/multi_node_config.yaml' &
 ```
 
 Once the workers are launched, you should see the output similar to the following in the worker logs.
@@ -323,7 +323,7 @@ Once the workers are launched, you should see the output similar to the followin
 
 4. Launch the router from node1 or login node.
 ```bash
-srun --mpi pmix -N 1 --ntasks 1 --ntasks-per-node=1 --overlap --container-image IMAGE --output batch_router_%x_%j.log --err batch_router_%x_%j.err --container-mounts PATH_TO_TRITON_DISTRIBUTED:/workspace  --container-env=NATS_SERVER,ETCD_ENDPOINTS bash -c 'cd /workspace/examples/python_rs/llm/tensorrt_llm && python3 -m disaggregated.router' &
+srun --mpi pmix -N 1 --ntasks 1 --ntasks-per-node=1 --overlap --container-image IMAGE --output batch_router_%x_%j.log --err batch_router_%x_%j.err --container-mounts PATH_TO_DYNAMO:/workspace  --container-env=NATS_SERVER,ETCD_ENDPOINTS bash -c 'cd /workspace/examples/python_rs/llm/tensorrt_llm && python3 -m disaggregated.router' &
 ```
 
 5. Send requests to the router.

examples/python_rs/llm/vllm/README.md

@@ -15,9 +15,9 @@ See the License for the specific language governing permissions and
 limitations under the License.
 -->
 
-# vLLM Integration with Triton Distributed
+# vLLM Integration with Dynamo
 
-This example demonstrates how to use Triton Distributed to serve large language models with the vLLM engine, enabling efficient model serving with both monolithic and disaggregated deployment options.
+This example demonstrates how to use Dynamo to serve large language models with the vLLM engine, enabling efficient model serving with both monolithic and disaggregated deployment options.
 
 ## Prerequisites
 
@@ -38,7 +38,7 @@ Start required services (etcd and NATS):
 
 ## Building the Environment
 
-The example is designed to run in a containerized environment using Triton Distributed, vLLM, and associated dependencies. To build the container:
+The example is designed to run in a containerized environment using Dynamo, vLLM, and associated dependencies. To build the container:
 
 ```bash
 # Build image

lib/bindings/python/README.md

@@ -15,9 +15,9 @@ See the License for the specific language governing permissions and
 limitations under the License.
 -->
 
-# Triton Distributed Python Bindings
+# Dynamo Python Bindings
 
-Python bindings for the Triton distributed runtime system, enabling distributed computing capabilities for machine learning workloads.
+Python bindings for the Dynamo runtime system, enabling distributed computing capabilities for machine learning workloads.
 
 ## 🚀 Quick Start
 
@@ -56,7 +56,7 @@ See [README.md](/lib/runtime/README.md).
 
 1. Start 3 separate shells, and activate the virtual environment in each
 ```
-cd python-wheels/triton-distributed
+cd python-wheels/dynamo
 source .venv/bin/activate
 ```
 

lib/runtime/README.md

@@ -15,13 +15,13 @@ See the License for the specific language governing permissions and
 limitations under the License.
 -->
 
-# Triton Distributed Runtime
+# Dynamo Runtime
 
 <h4>A Datacenter Scale Distributed Inference Serving Framework</h4>
 
 [![License](https://img.shields.io/badge/License-Apache_2.0-blue.svg)](https://opensource.org/licenses/Apache-2.0)
 
-Rust implementation of the Triton distributed runtime system, enabling distributed computing capabilities for machine learning workloads.
+Rust implementation of the Dynamo runtime system, enabling distributed computing capabilities for machine learning workloads.
 
 ## 🛠️ Prerequisites