Merge branch 'main' into ibhosale_grpc_streaming

CMakeLists.txt

@@ -38,22 +38,19 @@ set(CMAKE_CXX_STANDARD_REQUIRED ON)
 # Options
 #
 set(TRITON_VERSION "0.0.0" CACHE STRING "Version for the clients")
-set(PERF_ANALYZER_VERSION ${TRITON_VERSION} CACHE STRING "Build Version for Perf Analyzer")
 option(TRITON_ENABLE_CC_HTTP "Build C++ HTTP client libraries" OFF)
 option(TRITON_ENABLE_CC_GRPC "Build C++ GRPC client libraries" OFF)
 option(TRITON_ENABLE_PYTHON_HTTP "Enable Python HTTP client libraries" OFF)
 option(TRITON_ENABLE_PYTHON_GRPC "Enable Python GRPC client libraries" OFF)
 option(TRITON_ENABLE_JAVA_HTTP "Enable JAVA HTTP client libraries" OFF)
-option(TRITON_ENABLE_PERF_ANALYZER "Enable Performance Analyzer" OFF)
-option(TRITON_ENABLE_PERF_ANALYZER_C_API "Enable Performance Analyzer C API" OFF)
-option(TRITON_ENABLE_PERF_ANALYZER_TFS "Enable TensorFlow Serving support for Performance Analyzer" OFF)
-option(TRITON_ENABLE_PERF_ANALYZER_TS "Enable TorchServe support for Performance Analyzer" OFF)
-option(TRITON_ENABLE_PERF_ANALYZER_OPENAI "Enable OpenAI support for Performance Analyzer" OFF)
 option(TRITON_ENABLE_EXAMPLES "Include examples in build" OFF)
 option(TRITON_ENABLE_TESTS "Include tests in build" OFF)
 option(TRITON_ENABLE_GPU "Enable GPU support in libraries" OFF)
 option(TRITON_ENABLE_ZLIB "Include ZLIB library in build" ON)
 
+# Package Perf Analyzer with the Python Client -- Intended to be used by PA repo
+option(TRITON_PACKAGE_PERF_ANALYZER "Include Perf Analyzer in pip wheel" OFF)
+
 set(TRITON_REPO_ORGANIZATION "https://github.com/triton-inference-server" CACHE STRING "Git repository to pull from")
 set(TRITON_COMMON_REPO_TAG "main" CACHE STRING "Tag for triton-inference-server/common repo")
 set(TRITON_THIRD_PARTY_REPO_TAG "main" CACHE STRING "Tag for triton-inference-server/third_party repo")
@@ -128,28 +125,15 @@ else()
   set(_FINDPACKAGE_PROTOBUF_CONFIG_DIR "${TRITON_THIRD_PARTY_INSTALL_PREFIX}/protobuf/${LIB_DIR}/cmake/protobuf")
 endif()
 
-if(TRITON_ENABLE_CC_HTTP OR TRITON_ENABLE_CC_GRPC OR TRITON_ENABLE_PERF_ANALYZER OR TRITON_ENABLE_PERF_ANALYZER_C_API)
+if(TRITON_ENABLE_CC_HTTP OR TRITON_ENABLE_CC_GRPC)
 
   set(_cc_client_depends re2)
   if(${TRITON_ENABLE_CC_HTTP})
     set(_cc_client_depends ${_cc_client_depends} curl)
   endif() # TRITON_ENABLE_CC_HTTP
-  if(${TRITON_ENABLE_CC_GRPC} OR ${TRITON_ENABLE_PERF_ANALYZER})
+  if(${TRITON_ENABLE_CC_GRPC})
     set(_cc_client_depends ${_cc_client_depends} grpc protobuf)
-  endif() # TRITON_ENABLE_CC_GRPC OR TRITON_ENABLE_PERF_ANALYZER
-
-  if(NOT ${TRITON_ENABLE_PERF_ANALYZER} AND ${TRITON_ENABLE_PERF_ANALYZER_C_API})
-    message(FATAL_ERROR "TRITON_ENABLE_PERF_ANALYZER_C_API=ON requires TRITON_ENABLE_PERF_ANALYZER=ON")
-  endif() # NOT TRITON_ENABLE_PERF_ANALYZER AND TRITON_ENABLE_PERF_ANALYZER_C_API
-  if(NOT ${TRITON_ENABLE_PERF_ANALYZER} AND ${TRITON_ENABLE_PERF_ANALYZER_TFS})
-    message(FATAL_ERROR "TRITON_ENABLE_PERF_ANALYZER_TFS=ON requires TRITON_ENABLE_PERF_ANALYZER=ON")
-  endif() # NOT TRITON_ENABLE_PERF_ANALYZER AND TRITON_ENABLE_PERF_ANALYZER_TFS
-  if(NOT ${TRITON_ENABLE_PERF_ANALYZER} AND ${TRITON_ENABLE_PERF_ANALYZER_TS})
-    message(FATAL_ERROR "TRITON_ENABLE_PERF_ANALYZER_TS=ON requires TRITON_ENABLE_PERF_ANALYZER=ON")
-  endif() # NOT TRITON_ENABLE_PERF_ANALYZER AND TRITON_ENABLE_PERF_ANALYZER_TS
-  if(NOT ${TRITON_ENABLE_PERF_ANALYZER} AND ${TRITON_ENABLE_PERF_ANALYZER_OPENAI})
-    message(FATAL_ERROR "TRITON_ENABLE_PERF_ANALYZER_OPENAI=ON requires TRITON_ENABLE_PERF_ANALYZER=ON")
-  endif() # NOT TRITON_ENABLE_PERF_ANALYZER AND TRITON_ENABLE_PERF_ANALYZER_OPENAI
+  endif() # TRITON_ENABLE_CC_GRPC
 
   ExternalProject_Add(cc-clients
     PREFIX cc-clients
@@ -169,14 +153,8 @@ if(TRITON_ENABLE_CC_HTTP OR TRITON_ENABLE_CC_GRPC OR TRITON_ENABLE_PERF_ANALYZER
       -DTRITON_REPO_ORGANIZATION:STRING=${TRITON_REPO_ORGANIZATION}
       -DTRITON_COMMON_REPO_TAG:STRING=${TRITON_COMMON_REPO_TAG}
       -DTRITON_CORE_REPO_TAG:STRING=${TRITON_CORE_REPO_TAG}
-      -DPERF_ANALYZER_VERSION:STRING=${PERF_ANALYZER_VERSION}
       -DTRITON_ENABLE_CC_HTTP:BOOL=${TRITON_ENABLE_CC_HTTP}
       -DTRITON_ENABLE_CC_GRPC:BOOL=${TRITON_ENABLE_CC_GRPC}
-      -DTRITON_ENABLE_PERF_ANALYZER:BOOL=${TRITON_ENABLE_PERF_ANALYZER}
-      -DTRITON_ENABLE_PERF_ANALYZER_C_API:BOOL=${TRITON_ENABLE_PERF_ANALYZER_C_API}
-      -DTRITON_ENABLE_PERF_ANALYZER_TFS:BOOL=${TRITON_ENABLE_PERF_ANALYZER_TFS}
-      -DTRITON_ENABLE_PERF_ANALYZER_TS:BOOL=${TRITON_ENABLE_PERF_ANALYZER_TS}
-      -DTRITON_ENABLE_PERF_ANALYZER_OPENAI:BOOL=${TRITON_ENABLE_PERF_ANALYZER_OPENAI}
       -DTRITON_ENABLE_EXAMPLES:BOOL=${TRITON_ENABLE_EXAMPLES}
       -DTRITON_ENABLE_TESTS:BOOL=${TRITON_ENABLE_TESTS}
       -DTRITON_ENABLE_GPU:BOOL=${TRITON_ENABLE_GPU}
@@ -186,16 +164,13 @@ if(TRITON_ENABLE_CC_HTTP OR TRITON_ENABLE_CC_GRPC OR TRITON_ENABLE_PERF_ANALYZER
       -DCMAKE_INSTALL_PREFIX:PATH=${TRITON_INSTALL_PREFIX}
     DEPENDS ${_cc_client_depends}
   )
-endif() # TRITON_ENABLE_CC_HTTP OR TRITON_ENABLE_CC_GRPC OR TRITON_ENABLE_PERF_ANALYZER
+endif() # TRITON_ENABLE_CC_HTTP OR TRITON_ENABLE_CC_GRPC
 
 if(TRITON_ENABLE_PYTHON_HTTP OR TRITON_ENABLE_PYTHON_GRPC)
   set(_py_client_depends re2)
   if(${TRITON_ENABLE_PYTHON_GRPC})
     set(_py_client_depends ${_py_client_depends} grpc protobuf)
   endif() # TRITON_ENABLE_PYTHON_GRPC
-  if(${TRITON_ENABLE_PERF_ANALYZER})
-    set(_py_client_depends ${_py_client_depends} cc-clients)
-  endif() # TRITON_ENABLE_PERF_ANALYZER
 
   ExternalProject_Add(python-clients
     PREFIX python-clients
@@ -216,13 +191,9 @@ if(TRITON_ENABLE_PYTHON_HTTP OR TRITON_ENABLE_PYTHON_GRPC)
       -DTRITON_VERSION:STRING=${TRITON_VERSION}
       -DTRITON_ENABLE_PYTHON_HTTP:BOOL=${TRITON_ENABLE_PYTHON_HTTP}
       -DTRITON_ENABLE_PYTHON_GRPC:BOOL=${TRITON_ENABLE_PYTHON_GRPC}
-      -DTRITON_ENABLE_PERF_ANALYZER:BOOL=${TRITON_ENABLE_PERF_ANALYZER}
-      -DTRITON_ENABLE_PERF_ANALYZER_C_API:BOOL=${TRITON_ENABLE_PERF_ANALYZER_C_API}
-      -DTRITON_ENABLE_PERF_ANALYZER_TFS:BOOL=${TRITON_ENABLE_PERF_ANALYZER_TFS}
-      -DTRITON_ENABLE_PERF_ANALYZER_TS:BOOL=${TRITON_ENABLE_PERF_ANALYZER_TS}
-      -DTRITON_ENABLE_PERF_ANALYZER_OPENAI:BOOL=${TRITON_ENABLE_PERF_ANALYZER_OPENAI}
       -DTRITON_ENABLE_EXAMPLES:BOOL=${TRITON_ENABLE_EXAMPLES}
       -DTRITON_ENABLE_TESTS:BOOL=${TRITON_ENABLE_TESTS}
+      -DTRITON_PACKAGE_PERF_ANALYZER:BOOL=${TRITON_PACKAGE_PERF_ANALYZER}
       -DTRITON_ENABLE_GPU:BOOL=${TRITON_ENABLE_GPU}
       -DCMAKE_BUILD_TYPE:STRING=${CMAKE_BUILD_TYPE}
       -DCMAKE_INSTALL_PREFIX:PATH=${TRITON_INSTALL_PREFIX}

README.md

@@ -146,24 +146,9 @@ The components of the install packages are:
 * grpc [ `service_pb2`, `service_pb2_grpc`, `model_config_pb2` ]
 * utils [ linux distribution will include `shared_memory` and `cuda_shared_memory`]
 
-The Linux version of the package also includes the
-[perf_analyzer](src/c++/perf_analyzer/README.md)
-binary. The perf_analyzer binary is built on Ubuntu 20.04 and may not
-run on other Linux distributions. To run the perf_analyzer the
-following dependency must be installed:
-
-```bash
-$ sudo apt update
-$ sudo apt install libb64-dev
-```
-
-To reiterate, the installation on windows will not include perf_analyzer
-nor shared_memory/cuda_shared_memory components.
-
 ### Download From GitHub
 
-The client libraries and the perf_analyzer executable can be
-downloaded from the [Triton GitHub release
+The client libraries can be downloaded from the [Triton GitHub release
 page](https://github.com/triton-inference-server/server/releases)
 corresponding to the release you are interested in. The client
 libraries are found in the "Assets" section of the release page in a
@@ -186,15 +171,6 @@ include/, the Python wheel files in python/, and the jar files in
 java/.  The bin/ and python/ directories contain the built examples
 that you can learn more about below.
 
-The perf_analyzer binary is built on Ubuntu 20.04 and may not run on
-other Linux distributions. To use the C++ libraries or perf_analyzer
-executable you must install some dependencies.
-
-```bash
-$ apt-get update
-$ apt-get install curl libcurl4-openssl-dev libb64-dev
-```
-
 ### Download Docker Image From NGC
 
 A Docker image containing the client libraries and examples is
@@ -254,17 +230,6 @@ because Triton on Windows does not yet support all the build options.
 
 Use *cmake* to configure the build. You should adjust the flags depending on
 the components of Triton Client you are working and would like to build.
-For example, if you want to build Perf Analyzer with Triton C API, you can use \
-`-DTRITON_ENABLE_PERF_ANALYZER=ON -DTRITON_ENABLE_PERF_ANALYZER_C_API=ON`. You can
-also use `TRITON_ENABLE_PERF_ANALYZER_TFS` and `TRITON_ENABLE_PERF_ANALYZER_TS` flags
-to enable/disable support for TensorFlow Serving and TorchServe backend respectively in perf analyzer. \
-The following command demonstrate how to build client with all the features:
-
-```
-$ mkdir build
-$ cd build
-$ cmake -DCMAKE_INSTALL_PREFIX=`pwd`/install -DTRITON_ENABLE_CC_HTTP=ON -DTRITON_ENABLE_CC_GRPC=ON -DTRITON_ENABLE_PERF_ANALYZER=ON -DTRITON_ENABLE_PERF_ANALYZER_C_API=ON -DTRITON_ENABLE_PERF_ANALYZER_TFS=ON -DTRITON_ENABLE_PERF_ANALYZER_TS=ON -DTRITON_ENABLE_PYTHON_HTTP=ON -DTRITON_ENABLE_PYTHON_GRPC=ON -DTRITON_ENABLE_JAVA_HTTP=ON -DTRITON_ENABLE_GPU=ON -DTRITON_ENABLE_EXAMPLES=ON -DTRITON_ENABLE_TESTS=ON ..
-```
 
 If you are building on a release branch (or on a development branch
 that is based off of a release branch), then you must also use

src/c++/CMakeLists.txt

@@ -39,7 +39,6 @@ set(CMAKE_CXX_STANDARD_REQUIRED ON)
 #
 option(TRITON_ENABLE_CC_HTTP "Build C++ HTTP client libraries" OFF)
 option(TRITON_ENABLE_CC_GRPC "Build C++ GRPC client libraries" OFF)
-option(TRITON_ENABLE_PERF_ANALYZER "Enable Performance Analyzer" OFF)
 option(TRITON_ENABLE_EXAMPLES "Include examples in build" OFF)
 option(TRITON_ENABLE_TESTS "Include tests in build" OFF)
 option(TRITON_ENABLE_GPU "Enable GPU support in libraries" OFF)
@@ -71,26 +70,16 @@ FetchContent_Declare(
   URL https://github.com/google/googletest/archive/9406a60c7839052e4944ea4dbc8344762a89f9bd.zip
 )
 
-if(TRITON_ENABLE_CC_GRPC OR TRITON_ENABLE_PERF_ANALYZER)
+if(TRITON_ENABLE_CC_GRPC)
   set(TRITON_COMMON_ENABLE_PROTOBUF ON)
   set(TRITON_COMMON_ENABLE_GRPC ON)
+endif() # TRITON_ENABLE_CC_GRPC
 
-  if(TRITON_ENABLE_PERF_ANALYZER)
-    FetchContent_Declare(
-      repo-core
-      GIT_REPOSITORY ${TRITON_REPO_ORGANIZATION}/core.git
-      GIT_TAG ${TRITON_CORE_REPO_TAG}
-      GIT_SHALLOW ON
-    )
-    FetchContent_MakeAvailable(repo-core)
-  endif() # TRITON_ENABLE_PERF_ANALYZER
-endif() # TRITON_ENABLE_CC_GRPC OR TRITON_ENABLE_PERF_ANALYZER
-
-if(NOT TRITON_ENABLE_PERF_ANALYZER AND NOT TRITON_ENABLE_CC_HTTP AND NOT TRITON_ENABLE_EXAMPLES)
+if(NOT TRITON_ENABLE_CC_HTTP AND NOT TRITON_ENABLE_EXAMPLES)
   set(TRITON_COMMON_ENABLE_JSON OFF)
 endif()
 
-if(TRITON_ENABLE_TESTS OR TRITON_ENABLE_PERF_ANALYZER)
+if(TRITON_ENABLE_TESTS)
   FetchContent_MakeAvailable(googletest)
 endif()
 FetchContent_MakeAvailable(repo-common)
@@ -111,33 +100,33 @@ endif() # TRITON_ENABLE_GPU
 #
 # libcurl
 #
-if(TRITON_ENABLE_CC_HTTP OR TRITON_ENABLE_PERF_ANALYZER)
+if(TRITON_ENABLE_CC_HTTP)
   find_package(CURL REQUIRED)
   message(STATUS "Using curl ${CURL_VERSION}")
-endif() # TRITON_ENABLE_CC_HTTP OR TRITON_ENABLE_PERF_ANALYZER
+endif() # TRITON_ENABLE_CC_HTTP
 
 #
 # Protobuf
 #
-if(TRITON_ENABLE_CC_GRPC OR TRITON_ENABLE_PERF_ANALYZER)
+if(TRITON_ENABLE_CC_GRPC)
   set(protobuf_MODULE_COMPATIBLE TRUE CACHE BOOL "protobuf_MODULE_COMPATIBLE" FORCE)
   find_package(Protobuf CONFIG REQUIRED)
   message(STATUS "Using protobuf ${Protobuf_VERSION}")
   include_directories(${Protobuf_INCLUDE_DIRS})
-endif() # TRITON_ENABLE_CC_GRPC OR TRITON_ENABLE_PERF_ANALYZER
+endif() # TRITON_ENABLE_CC_GRPC
 
 #
 # GRPC
 #
-if(TRITON_ENABLE_CC_GRPC OR TRITON_ENABLE_PERF_ANALYZER)
+if(TRITON_ENABLE_CC_GRPC)
   find_package(gRPC CONFIG REQUIRED)
   message(STATUS "Using gRPC ${gRPC_VERSION}")
   include_directories($<TARGET_PROPERTY:gRPC::grpc,INTERFACE_INCLUDE_DIRECTORIES>)
-endif() # TRITON_ENABLE_CC_GRPC OR TRITON_ENABLE_PERF_ANALYZER
+endif() # TRITON_ENABLE_CC_GRPC
 
-if(TRITON_ENABLE_CC_HTTP OR TRITON_ENABLE_CC_GRPC OR TRITON_ENABLE_PERF_ANALYZER)
+if(TRITON_ENABLE_CC_HTTP OR TRITON_ENABLE_CC_GRPC)
   add_subdirectory(library)
-endif() # TRITON_ENABLE_CC_HTTP OR TRITON_ENABLE_CC_GRPC OR TRITON_ENABLE_PERF_ANALYZER
+endif() # TRITON_ENABLE_CC_HTTP OR TRITON_ENABLE_CC_GRPC
 
 if(TRITON_ENABLE_CC_HTTP OR TRITON_ENABLE_CC_GRPC)
   if(TRITON_ENABLE_EXAMPLES)
@@ -148,7 +137,3 @@ if(TRITON_ENABLE_CC_HTTP OR TRITON_ENABLE_CC_GRPC)
     add_subdirectory(tests)
   endif() # TRITON_ENABLE_TESTS
 endif() # TRITON_ENABLE_CC_HTTP OR TRITON_ENABLE_CC_GRPC
-
-if(TRITON_ENABLE_PERF_ANALYZER)
-  add_subdirectory(perf_analyzer)
-endif() # TRITON_ENABLE_PERF_ANALYZER