Skip to content

Commit 48cc33c

Browse files
dzierdzier
authored
Update README for 21.08 release (#3273)
1 parent f904435 commit 48cc33c

File tree

1 file changed

+265
-2
lines changed

1 file changed

+265
-2
lines changed

‎README.md

+265-2
Original file line numberDiff line numberDiff line change
@@ -30,5 +30,268 @@
3030

3131
# Triton Inference Server
3232

33-
**NOTE: You are currently on the r21.08 branch which tracks stabilization
34-
towards the next release. This branch is not usable during stabilization.**
33+
Triton Inference Server provides a cloud and edge inferencing solution
34+
optimized for both CPUs and GPUs. Triton supports an HTTP/REST and
35+
GRPC protocol that allows remote clients to request inferencing for
36+
any model being managed by the server. For edge deployments, Triton is
37+
available as a shared library with a C API that allows the full
38+
functionality of Triton to be included directly in an
39+
application.
40+
41+
## What's New in 2.13.0
42+
43+
* Initial Beta release for [Business Logic Scripting](https://github.com/triton-inference-server/python_backend/tree/r21.08#business-logic-scripting-beta),
44+
a new set of utility functions that allow the execution of inference requests
45+
on other models being served by Triton as part of executing a Python model.
46+
47+
* Release new [Container Composition Utility](docs/compose.md) which can be used
48+
to create custom Triton containers with specific backends and repository
49+
agents.
50+
51+
* Starting in 21.08, Triton will release two new containers on NGC.
52+
* nvcr.io/nvidia/tritonserver:21.08-tf-python-py3 - GPU enabled Triton server
53+
with only the TensorFlow 2.x and Python backends.
54+
* nvcr.io/nvidia/tritonserver:21.08-pyt-python-py3 - GPU enabled Triton server
55+
with only the PyTorch and Python backends.
56+
57+
* Added Model Analyzer support for models with custom operations.
58+
59+
## Features
60+
61+
* [Multiple deep-learning
62+
frameworks](https://github.com/triton-inference-server/backend). Triton
63+
can manage any number and mix of models (limited by system disk and
64+
memory resources). Triton supports TensorRT, TensorFlow GraphDef,
65+
TensorFlow SavedModel, ONNX, PyTorch TorchScript and OpenVINO model
66+
formats. Both TensorFlow 1.x and TensorFlow 2.x are
67+
supported. Triton also supports TensorFlow-TensorRT and
68+
ONNX-TensorRT integrated models.
69+
70+
* [Concurrent model
71+
execution](docs/architecture.md#concurrent-model-execution). Multiple
72+
models (or multiple instances of the same model) can run
73+
simultaneously on the same GPU or on multiple GPUs.
74+
75+
* [Dynamic batching](docs/architecture.md#models-and-schedulers). For
76+
models that support batching, Triton implements multiple scheduling
77+
and batching algorithms that combine individual inference requests
78+
together to improve inference throughput. These scheduling and
79+
batching decisions are transparent to the client requesting
80+
inference.
81+
82+
* [Extensible
83+
backends](https://github.com/triton-inference-server/backend). In
84+
addition to deep-learning frameworks, Triton provides a *backend
85+
API* that allows Triton to be extended with any model execution
86+
logic implemented in
87+
[Python](https://github.com/triton-inference-server/python_backend)
88+
or
89+
[C++](https://github.com/triton-inference-server/backend/blob/main/README.md#triton-backend-api),
90+
while still benefiting from the CPU and GPU support, concurrent
91+
execution, dynamic batching and other features provided by Triton.
92+
93+
* [Model pipelines](docs/architecture.md#ensemble-models). Triton
94+
*ensembles* represents a pipeline of one or more models and the
95+
connection of input and output tensors between those models. A
96+
single inference request to an ensemble will trigger the execution
97+
of the entire pipeline.
98+
99+
* [HTTP/REST and GRPC inference
100+
protocols](docs/inference_protocols.md) based on the community
101+
developed [KFServing
102+
protocol](https://github.com/kubeflow/kfserving/tree/master/docs/predict-api/v2).
103+
104+
* A [C API](docs/inference_protocols.md#c-api) allows Triton to be
105+
linked directly into your application for edge and other in-process
106+
use cases.
107+
108+
* [Metrics](docs/metrics.md) indicating GPU utilization, server
109+
throughput, and server latency. The metrics are provided in
110+
Prometheus data format.
111+
112+
## Documentation
113+
114+
[Triton Architecture](docs/architecture.md) gives a high-level
115+
overview of the structure and capabilities of the inference
116+
server. There is also an [FAQ](docs/faq.md). Additional documentation
117+
is divided into [*user*](#user-documentation) and
118+
[*developer*](#developer-documentation) sections. The *user*
119+
documentation describes how to use Triton as an inference solution,
120+
including information on how to configure Triton, how to organize and
121+
configure your models, how to use the C++ and Python clients, etc. The
122+
*developer* documentation describes how to build and test Triton and
123+
also how Triton can be extended with new functionality.
124+
125+
The Triton [Release
126+
Notes](https://docs.nvidia.com/deeplearning/triton-inference-server/release-notes/index.html)
127+
and [Support
128+
Matrix](https://docs.nvidia.com/deeplearning/dgx/support-matrix/index.html)
129+
indicate the required versions of the NVIDIA Driver and CUDA, and also
130+
describe supported GPUs.
131+
132+
### User Documentation
133+
134+
- [QuickStart](docs/quickstart.md)
135+
- [Install](docs/quickstart.md#install-triton-docker-image)
136+
- [Run](docs/quickstart.md#run-triton)
137+
- [Model Repository](docs/model_repository.md)
138+
- [Model Configuration](docs/model_configuration.md)
139+
- [Model Management](docs/model_management.md)
140+
- [Custom Operations](docs/custom_operations.md)
141+
- [Client Libraries and Examples](https://github.com/triton-inference-server/client)
142+
- [Optimization](docs/optimization.md)
143+
- [Model Analyzer](docs/model_analyzer.md)
144+
- [Performance Analyzer](docs/perf_analyzer.md)
145+
- [Metrics](docs/metrics.md)
146+
- [Jetson and JetPack](docs/jetson.md)
147+
148+
The [quickstart](docs/quickstart.md) walks you through all the steps
149+
required to install and run Triton with an example image
150+
classification model and then use an example client application to
151+
perform inferencing using that model. The quickstart also demonstrates
152+
how [Triton supports both GPU systems and CPU-only
153+
systems](docs/quickstart.md#run-triton).
154+
155+
The first step in using Triton to serve your models is to place one or
156+
more models into a [model
157+
repository](docs/model_repository.md). Optionally, depending on the type
158+
of the model and on what Triton capabilities you want to enable for
159+
the model, you may need to create a [model
160+
configuration](docs/model_configuration.md) for the model. If your
161+
model has [custom operations](docs/custom_operations.md) you will need
162+
to make sure they are loaded correctly by Triton.
163+
164+
After you have your model(s) available in Triton, you will want to
165+
send inference and other requests to Triton from your *client*
166+
application. The [Python and C++ client
167+
libraries](https://github.com/triton-inference-server/client) provide
168+
APIs to simplify this communication. There are also a large number of
169+
[client examples](https://github.com/triton-inference-server/client)
170+
that demonstrate how to use the libraries. You can also send
171+
HTTP/REST requests directly to Triton using the [HTTP/REST JSON-based
172+
protocol](docs/inference_protocols.md#httprest-and-grpc-protocols) or
173+
[generate a GRPC client for many other
174+
languages](https://github.com/triton-inference-server/client).
175+
176+
Understanding and [optimizing performance](docs/optimization.md) is an
177+
important part of deploying your models. The Triton project provides
178+
the [Performance Analyzer](docs/perf_analyzer.md) and the [Model
179+
Analyzer](docs/model_analyzer.md) to help your optimization
180+
efforts. Specifically, you will want to optimize [scheduling and
181+
batching](docs/architecture.md#models-and-schedulers) and [model
182+
instances](docs/model_configuration.md#instance-groups) appropriately
183+
for each model. You may also want to consider [ensembling multiple
184+
models and pre/post-processing](docs/architecture.md#ensemble-models)
185+
into a pipeline. In some cases you may find [individual inference
186+
request trace data](docs/trace.md) useful when optimizing. A
187+
[Prometheus metrics endpoint](docs/metrics.md) allows you to visualize
188+
and monitor aggregate inference metrics.
189+
190+
NVIDIA publishes a number of [deep learning
191+
examples](https://github.com/NVIDIA/DeepLearningExamples) that use
192+
Triton.
193+
194+
As part of your deployment strategy you may want to [explicitly manage
195+
what models are available by loading and unloading
196+
models](docs/model_management.md) from a running Triton server. If you
197+
are using Kubernetes for deployment there are simple examples of how
198+
to deploy Triton using Kubernetes and Helm, one for
199+
[GCP](deploy/gcp/README.md) and one for [AWS](deploy/aws/README.md).
200+
201+
The [version 1 to version 2 migration
202+
information](docs/v1_to_v2.md) is helpful if you are moving to
203+
version 2 of Triton from previously using version 1.
204+
205+
### Developer Documentation
206+
207+
- [Build](docs/build.md)
208+
- [Protocols and APIs](docs/inference_protocols.md).
209+
- [Backends](https://github.com/triton-inference-server/backend)
210+
- [Repository Agents](docs/repository_agents.md)
211+
- [Test](docs/test.md)
212+
213+
Triton can be [built using
214+
Docker](docs/build.md#building-triton-with-docker) or [built without
215+
Docker](docs/build.md#building-triton-without-docker). After building
216+
you should [test Triton](docs/test.md).
217+
218+
It is also possible to [create a Docker image containing a customized
219+
Triton](docs/compose.md) that contains only a subset of the backends.
220+
221+
The Triton project also provides [client libraries for Python and
222+
C++](https://github.com/triton-inference-server/client) that make it
223+
easy to communicate with the server. There are also a large number of
224+
[example clients](https://github.com/triton-inference-server/client)
225+
that demonstrate how to use the libraries. You can also develop your
226+
own clients that directly communicate with Triton using [HTTP/REST or
227+
GRPC protocols](docs/inference_protocols.md). There is also a [C
228+
API](docs/inference_protocols.md) that allows Triton to be linked
229+
directly into your application.
230+
231+
A [Triton backend](https://github.com/triton-inference-server/backend)
232+
is the implementation that executes a model. A backend can interface
233+
with a deep learning framework, like PyTorch, TensorFlow, TensorRT or
234+
ONNX Runtime; or it can interface with a data processing framework
235+
like [DALI](https://github.com/triton-inference-server/dali_backend);
236+
or you can extend Triton by [writing your own
237+
backend](https://github.com/triton-inference-server/backend) in either
238+
[C/C++](https://github.com/triton-inference-server/backend/blob/main/README.md#triton-backend-api)
239+
or
240+
[Python](https://github.com/triton-inference-server/python_backend).
241+
242+
A [Triton repository agent](docs/repository_agents.md) extends Triton
243+
with new functionality that operates when a model is loaded or
244+
unloaded. You can introduce your own code to perform authentication,
245+
decryption, conversion, or similar operations when a model is loaded.
246+
247+
## Papers and Presentation
248+
249+
* [Maximizing Deep Learning Inference Performance with NVIDIA Model
250+
Analyzer](https://developer.nvidia.com/blog/maximizing-deep-learning-inference-performance-with-nvidia-model-analyzer/).
251+
252+
* [High-Performance Inferencing at Scale Using the TensorRT Inference
253+
Server](https://developer.nvidia.com/gtc/2020/video/s22418).
254+
255+
* [Accelerate and Autoscale Deep Learning Inference on GPUs with
256+
KFServing](https://developer.nvidia.com/gtc/2020/video/s22459).
257+
258+
* [Deep into Triton Inference Server: BERT Practical Deployment on
259+
NVIDIA GPU](https://developer.nvidia.com/gtc/2020/video/s21736).
260+
261+
* [Maximizing Utilization for Data Center Inference with TensorRT
262+
Inference Server](https://on-demand-gtc.gputechconf.com/gtcnew/sessionview.php?sessionName=s9438-maximizing+utilization+for+data+center+inference+with+tensorrt+inference+server).
263+
264+
* [NVIDIA TensorRT Inference Server Boosts Deep Learning
265+
Inference](https://devblogs.nvidia.com/nvidia-serves-deep-learning-inference/).
266+
267+
* [GPU-Accelerated Inference for Kubernetes with the NVIDIA TensorRT
268+
Inference Server and
269+
Kubeflow](https://www.kubeflow.org/blog/nvidia_tensorrt/).
270+
271+
## Contributing
272+
273+
Contributions to Triton Inference Server are more than welcome. To
274+
contribute make a pull request and follow the guidelines outlined in
275+
[CONTRIBUTING.md](CONTRIBUTING.md). If you have a backend, client,
276+
example or similar contribution that is not modifying the core of
277+
Triton, then you should file a PR in the [contrib
278+
repo](https://github.com/triton-inference-server/contrib).
279+
280+
## Reporting problems, asking questions
281+
282+
We appreciate any feedback, questions or bug reporting regarding this
283+
project. When help with code is needed, follow the process outlined in
284+
the Stack Overflow (https://stackoverflow.com/help/mcve)
285+
document. Ensure posted examples are:
286+
287+
* minimal – use as little code as possible that still produces the
288+
same problem
289+
290+
* complete – provide all parts needed to reproduce the problem. Check
291+
if you can strip external dependency and still show the problem. The
292+
less time we spend on reproducing problems the more time we have to
293+
fix it
294+
295+
* verifiable – test the code you're about to provide to make sure it
296+
reproduces the problem. Remove all other problems that are not
297+
related to your request/question.

0 commit comments

Comments
 (0)