diff --git a/BUILD.md b/BUILD.md
index db5ca3037f86c..98c1684de5b21 100644
--- a/BUILD.md
+++ b/BUILD.md
@@ -53,9 +53,9 @@ The complete list of build options can be found by running `./build.sh (or ./bui
 1. For Windows, just add --x86 argument when launching build.bat
 2. For Linux, it must be built out of a x86 os, --x86 argument also needs be specified to build.sh
 
-## Build Hosting Application on Linux
+## Build ONNX Runtime Server on Linux
 
-1. In the ONNX Runtime root folder, run `./build.sh --config RelWithDebInfo --build_hosting  --use_openmp --parallel`
+1. In the ONNX Runtime root folder, run `./build.sh --config RelWithDebInfo --build_server  --use_openmp --parallel`
 
 ## Build/Test Flavors for CI
 
diff --git a/cmake/CMakeLists.txt b/cmake/CMakeLists.txt
index c717e86b75136..0f5b3d85881ea 100644
--- a/cmake/CMakeLists.txt
+++ b/cmake/CMakeLists.txt
@@ -70,7 +70,7 @@ option(onnxruntime_USE_BRAINSLICE "Build with BrainSlice" OFF)
 option(onnxruntime_USE_TENSORRT "Build with TensorRT support" OFF)
 option(onnxruntime_ENABLE_LTO "Enable link time optimization, which is not stable on older GCCs" OFF)
 option(onnxruntime_CROSS_COMPILING "Cross compiling onnx runtime" OFF)
-option(onnxruntime_BUILD_HOSTING "Build ONNX hosting service" OFF)
+option(onnxruntime_BUILD_SERVER "Build ONNX Runtime Server" OFF)
 option(onnxruntime_USE_FULL_PROTOBUF "Use full protobuf" OFF)
 option(onnxruntime_DISABLE_CONTRIB_OPS "Disable contrib ops" OFF)
 option(onnxruntime_USE_EIGEN_THREADPOOL "Use eigen threadpool. Otherwise OpenMP or a homemade one will be used" OFF)
@@ -608,8 +608,8 @@ if (onnxruntime_BUILD_SHARED_LIB)
   include(onnxruntime.cmake)
 endif()
 
-if (onnxruntime_BUILD_HOSTING)
-  include(onnxruntime_hosting.cmake)
+if (onnxruntime_BUILD_SERVER)
+  include(onnxruntime_server.cmake)
 endif()
 
 # some of the tests rely on the shared libs to be
diff --git a/cmake/get_boost.cmake b/cmake/get_boost.cmake
index 5bcefa3479fd4..265d088698571 100644
--- a/cmake/get_boost.cmake
+++ b/cmake/get_boost.cmake
@@ -1,3 +1,6 @@
+# Copyright (c) Microsoft Corporation. All rights reserved.
+# Licensed under the MIT License.
+
 set(BOOST_REQUESTED_VERSION 1.69.0 CACHE STRING "")
 set(BOOST_SHA1 8f32d4617390d1c2d16f26a27ab60d97807b35440d45891fa340fc2648b04406 CACHE STRING "")
 set(BOOST_USE_STATIC_LIBS true CACHE BOOL "")
diff --git a/cmake/onnxruntime_hosting.cmake b/cmake/onnxruntime_hosting.cmake
deleted file mode 100644
index 691703ac58a45..0000000000000
--- a/cmake/onnxruntime_hosting.cmake
+++ /dev/null
@@ -1,120 +0,0 @@
-# Copyright (c) Microsoft Corporation. All rights reserved.
-# Licensed under the MIT License.
-
-set(HOSTING_APP_NAME "onnxruntime_hosting")
-
-# Generate .h and .cc files from protobuf file
-add_library(hosting_proto
-  ${ONNXRUNTIME_ROOT}/hosting/protobuf/predict.proto)
-target_include_directories(hosting_proto PUBLIC $<TARGET_PROPERTY:protobuf::libprotobuf,INTERFACE_INCLUDE_DIRECTORIES> "${CMAKE_CURRENT_BINARY_DIR}/.." ${CMAKE_CURRENT_BINARY_DIR}/onnx)
-target_compile_definitions(hosting_proto PUBLIC $<TARGET_PROPERTY:protobuf::libprotobuf,INTERFACE_COMPILE_DEFINITIONS>)
-onnxruntime_protobuf_generate(APPEND_PATH IMPORT_DIRS ${REPO_ROOT}/cmake/external/protobuf/src ${ONNXRUNTIME_ROOT}/hosting/protobuf ${ONNXRUNTIME_ROOT}/core/protobuf TARGET hosting_proto)
-add_dependencies(hosting_proto onnx_proto ${onnxruntime_EXTERNAL_DEPENDENCIES})
-if(NOT WIN32)
-  if(HAS_UNUSED_PARAMETER)
-    set_source_files_properties(${CMAKE_CURRENT_BINARY_DIR}/model_metadata.pb.cc PROPERTIES COMPILE_FLAGS -Wno-unused-parameter)
-    set_source_files_properties(${CMAKE_CURRENT_BINARY_DIR}/model_status.pb.cc PROPERTIES COMPILE_FLAGS -Wno-unused-parameter)
-    set_source_files_properties(${CMAKE_CURRENT_BINARY_DIR}/predict.pb.cc PROPERTIES COMPILE_FLAGS -Wno-unused-parameter)
-  endif()
-endif()
-
-# Setup dependencies
-include(get_boost.cmake)
-set(re2_src ${REPO_ROOT}/cmake/external/re2)
-
-# Setup source code
-set(onnxruntime_hosting_lib_srcs
-  "${ONNXRUNTIME_ROOT}/hosting/http/json_handling.cc"
-  "${ONNXRUNTIME_ROOT}/hosting/http/predict_request_handler.cc"
-  "${ONNXRUNTIME_ROOT}/hosting/http/util.cc"
-  "${ONNXRUNTIME_ROOT}/hosting/environment.cc"
-  "${ONNXRUNTIME_ROOT}/hosting/executor.cc"
-  "${ONNXRUNTIME_ROOT}/hosting/converter.cc"
-  "${ONNXRUNTIME_ROOT}/hosting/util.cc"
-  )
-if(NOT WIN32)
-  if(HAS_UNUSED_PARAMETER)
-    set_source_files_properties(${ONNXRUNTIME_ROOT}/hosting/http/json_handling.cc PROPERTIES COMPILE_FLAGS -Wno-unused-parameter)
-    set_source_files_properties(${ONNXRUNTIME_ROOT}/hosting/http/predict_request_handler.cc PROPERTIES COMPILE_FLAGS -Wno-unused-parameter)
-    set_source_files_properties(${ONNXRUNTIME_ROOT}/hosting/executor.cc PROPERTIES COMPILE_FLAGS -Wno-unused-parameter)
-    set_source_files_properties(${ONNXRUNTIME_ROOT}/hosting/converter.cc PROPERTIES COMPILE_FLAGS -Wno-unused-parameter)
-    set_source_files_properties(${ONNXRUNTIME_ROOT}/hosting/util.cc PROPERTIES COMPILE_FLAGS -Wno-unused-parameter)
-  endif()
-endif()
-
-file(GLOB_RECURSE onnxruntime_hosting_http_core_lib_srcs
-  "${ONNXRUNTIME_ROOT}/hosting/http/core/*.cc"
-  )
-
-file(GLOB_RECURSE onnxruntime_hosting_srcs
-  "${ONNXRUNTIME_ROOT}/hosting/main.cc"
-)
-
-# HTTP core library
-add_library(onnxruntime_hosting_http_core_lib STATIC
-  ${onnxruntime_hosting_http_core_lib_srcs})
-target_include_directories(onnxruntime_hosting_http_core_lib
-  PUBLIC
-  ${ONNXRUNTIME_ROOT}/hosting/http/core
-  ${Boost_INCLUDE_DIR}
-  ${re2_src}
-)
-add_dependencies(onnxruntime_hosting_http_core_lib Boost)
-target_link_libraries(onnxruntime_hosting_http_core_lib PRIVATE
-  ${Boost_LIBRARIES}
-)
-
-# Hosting library
-add_library(onnxruntime_hosting_lib ${onnxruntime_hosting_lib_srcs})
-onnxruntime_add_include_to_target(onnxruntime_hosting_lib gsl onnx_proto hosting_proto)
-target_include_directories(onnxruntime_hosting_lib PRIVATE
-  ${ONNXRUNTIME_ROOT}
-  ${CMAKE_CURRENT_BINARY_DIR}/onnx
-  ${ONNXRUNTIME_ROOT}/hosting
-  ${ONNXRUNTIME_ROOT}/hosting/http
-  PUBLIC
-  ${Boost_INCLUDE_DIR}
-  ${re2_src}
-)
-
-target_link_libraries(onnxruntime_hosting_lib PRIVATE
-  hosting_proto
-  ${Boost_LIBRARIES}
-  onnxruntime_hosting_http_core_lib
-  onnxruntime_session
-  onnxruntime_optimizer
-  onnxruntime_providers
-  onnxruntime_util
-  onnxruntime_framework
-  onnxruntime_util
-  onnxruntime_graph
-  onnxruntime_common
-  onnxruntime_mlas
-  ${onnxruntime_EXTERNAL_LIBRARIES}
-)
-
-# For IDE only
-source_group(TREE ${REPO_ROOT} FILES ${onnxruntime_hosting_srcs} ${onnxruntime_hosting_lib_srcs} ${onnxruntime_hosting_lib})
-
-# Hosting Application
-add_executable(${HOSTING_APP_NAME} ${onnxruntime_hosting_srcs})
-add_dependencies(${HOSTING_APP_NAME} onnx hosting_proto onnx_proto ${onnxruntime_EXTERNAL_DEPENDENCIES})
-
-if(NOT WIN32)
-  if(HAS_UNUSED_PARAMETER)
-    set_source_files_properties("${ONNXRUNTIME_ROOT}/hosting/main.cc" PROPERTIES COMPILE_FLAGS -Wno-unused-parameter)
-  endif()
-endif()
-
-onnxruntime_add_include_to_target(${HOSTING_APP_NAME} onnxruntime_session onnxruntime_hosting_lib gsl onnx onnx_proto hosting_proto)
-
-target_include_directories(${HOSTING_APP_NAME} PRIVATE
-    ${ONNXRUNTIME_ROOT}
-    ${ONNXRUNTIME_ROOT}/hosting/http
-)
-
-target_link_libraries(${HOSTING_APP_NAME} PRIVATE
-    onnxruntime_hosting_http_core_lib
-    onnxruntime_hosting_lib
-)
-
diff --git a/cmake/onnxruntime_server.cmake b/cmake/onnxruntime_server.cmake
new file mode 100644
index 0000000000000..f67df80e4da35
--- /dev/null
+++ b/cmake/onnxruntime_server.cmake
@@ -0,0 +1,120 @@
+# Copyright (c) Microsoft Corporation. All rights reserved.
+# Licensed under the MIT License.
+
+set(SERVER_APP_NAME "onnxruntime_server")
+
+# Generate .h and .cc files from protobuf file
+add_library(server_proto
+  ${ONNXRUNTIME_ROOT}/server/protobuf/predict.proto)
+target_include_directories(server_proto PUBLIC $<TARGET_PROPERTY:protobuf::libprotobuf,INTERFACE_INCLUDE_DIRECTORIES> "${CMAKE_CURRENT_BINARY_DIR}/.." ${CMAKE_CURRENT_BINARY_DIR}/onnx)
+target_compile_definitions(server_proto PUBLIC $<TARGET_PROPERTY:protobuf::libprotobuf,INTERFACE_COMPILE_DEFINITIONS>)
+onnxruntime_protobuf_generate(APPEND_PATH IMPORT_DIRS ${REPO_ROOT}/cmake/external/protobuf/src ${ONNXRUNTIME_ROOT}/server/protobuf ${ONNXRUNTIME_ROOT}/core/protobuf TARGET server_proto)
+add_dependencies(server_proto onnx_proto ${onnxruntime_EXTERNAL_DEPENDENCIES})
+if(NOT WIN32)
+  if(HAS_UNUSED_PARAMETER)
+    set_source_files_properties(${CMAKE_CURRENT_BINARY_DIR}/model_metadata.pb.cc PROPERTIES COMPILE_FLAGS -Wno-unused-parameter)
+    set_source_files_properties(${CMAKE_CURRENT_BINARY_DIR}/model_status.pb.cc PROPERTIES COMPILE_FLAGS -Wno-unused-parameter)
+    set_source_files_properties(${CMAKE_CURRENT_BINARY_DIR}/predict.pb.cc PROPERTIES COMPILE_FLAGS -Wno-unused-parameter)
+  endif()
+endif()
+
+# Setup dependencies
+include(get_boost.cmake)
+set(re2_src ${REPO_ROOT}/cmake/external/re2)
+
+# Setup source code
+set(onnxruntime_server_lib_srcs
+  "${ONNXRUNTIME_ROOT}/server/http/json_handling.cc"
+  "${ONNXRUNTIME_ROOT}/server/http/predict_request_handler.cc"
+  "${ONNXRUNTIME_ROOT}/server/http/util.cc"
+  "${ONNXRUNTIME_ROOT}/server/environment.cc"
+  "${ONNXRUNTIME_ROOT}/server/executor.cc"
+  "${ONNXRUNTIME_ROOT}/server/converter.cc"
+  "${ONNXRUNTIME_ROOT}/server/util.cc"
+  )
+if(NOT WIN32)
+  if(HAS_UNUSED_PARAMETER)
+    set_source_files_properties(${ONNXRUNTIME_ROOT}/server/http/json_handling.cc PROPERTIES COMPILE_FLAGS -Wno-unused-parameter)
+    set_source_files_properties(${ONNXRUNTIME_ROOT}/server/http/predict_request_handler.cc PROPERTIES COMPILE_FLAGS -Wno-unused-parameter)
+    set_source_files_properties(${ONNXRUNTIME_ROOT}/server/executor.cc PROPERTIES COMPILE_FLAGS -Wno-unused-parameter)
+    set_source_files_properties(${ONNXRUNTIME_ROOT}/server/converter.cc PROPERTIES COMPILE_FLAGS -Wno-unused-parameter)
+    set_source_files_properties(${ONNXRUNTIME_ROOT}/server/util.cc PROPERTIES COMPILE_FLAGS -Wno-unused-parameter)
+  endif()
+endif()
+
+file(GLOB_RECURSE onnxruntime_server_http_core_lib_srcs
+  "${ONNXRUNTIME_ROOT}/server/http/core/*.cc"
+  )
+
+file(GLOB_RECURSE onnxruntime_server_srcs
+  "${ONNXRUNTIME_ROOT}/server/main.cc"
+)
+
+# HTTP core library
+add_library(onnxruntime_server_http_core_lib STATIC
+  ${onnxruntime_server_http_core_lib_srcs})
+target_include_directories(onnxruntime_server_http_core_lib
+  PUBLIC
+  ${ONNXRUNTIME_ROOT}/server/http/core
+  ${Boost_INCLUDE_DIR}
+  ${re2_src}
+)
+add_dependencies(onnxruntime_server_http_core_lib Boost)
+target_link_libraries(onnxruntime_server_http_core_lib PRIVATE
+  ${Boost_LIBRARIES}
+)
+
+# Server library
+add_library(onnxruntime_server_lib ${onnxruntime_server_lib_srcs})
+onnxruntime_add_include_to_target(onnxruntime_server_lib gsl onnx_proto server_proto)
+target_include_directories(onnxruntime_server_lib PRIVATE
+  ${ONNXRUNTIME_ROOT}
+  ${CMAKE_CURRENT_BINARY_DIR}/onnx
+  ${ONNXRUNTIME_ROOT}/server
+  ${ONNXRUNTIME_ROOT}/server/http
+  PUBLIC
+  ${Boost_INCLUDE_DIR}
+  ${re2_src}
+)
+
+target_link_libraries(onnxruntime_server_lib PRIVATE
+  server_proto
+  ${Boost_LIBRARIES}
+  onnxruntime_server_http_core_lib
+  onnxruntime_session
+  onnxruntime_optimizer
+  onnxruntime_providers
+  onnxruntime_util
+  onnxruntime_framework
+  onnxruntime_util
+  onnxruntime_graph
+  onnxruntime_common
+  onnxruntime_mlas
+  ${onnxruntime_EXTERNAL_LIBRARIES}
+)
+
+# For IDE only
+source_group(TREE ${REPO_ROOT} FILES ${onnxruntime_server_srcs} ${onnxruntime_server_lib_srcs} ${onnxruntime_server_lib})
+
+# Server Application
+add_executable(${SERVER_APP_NAME} ${onnxruntime_server_srcs})
+add_dependencies(${SERVER_APP_NAME} onnx server_proto onnx_proto ${onnxruntime_EXTERNAL_DEPENDENCIES})
+
+if(NOT WIN32)
+  if(HAS_UNUSED_PARAMETER)
+    set_source_files_properties("${ONNXRUNTIME_ROOT}/server/main.cc" PROPERTIES COMPILE_FLAGS -Wno-unused-parameter)
+  endif()
+endif()
+
+onnxruntime_add_include_to_target(${SERVER_APP_NAME} onnxruntime_session onnxruntime_server_lib gsl onnx onnx_proto server_proto)
+
+target_include_directories(${SERVER_APP_NAME} PRIVATE
+    ${ONNXRUNTIME_ROOT}
+    ${ONNXRUNTIME_ROOT}/server/http
+)
+
+target_link_libraries(${SERVER_APP_NAME} PRIVATE
+    onnxruntime_server_http_core_lib
+    onnxruntime_server_lib
+)
+
diff --git a/cmake/onnxruntime_unittests.cmake b/cmake/onnxruntime_unittests.cmake
index 04db354db3c90..73b5b48be5fa9 100644
--- a/cmake/onnxruntime_unittests.cmake
+++ b/cmake/onnxruntime_unittests.cmake
@@ -162,9 +162,9 @@ set(onnxruntime_test_framework_libs
   onnxruntime_mlas
   )
 
-set(onnxruntime_test_hosting_libs
+set(onnxruntime_test_server_libs
   onnxruntime_test_utils
-  onnxruntime_test_utils_for_hosting
+  onnxruntime_test_utils_for_server
 )
 
 if(WIN32)
@@ -552,52 +552,55 @@ if (onnxruntime_BUILD_SHARED_LIB)
   endif()
 endif()
 
-if (onnxruntime_BUILD_HOSTING)
-  file(GLOB onnxruntime_test_hosting_src
-    "${TEST_SRC_DIR}/hosting/unit_tests/*.cc"
-    "${TEST_SRC_DIR}/hosting/unit_tests/*.h"
+if (onnxruntime_BUILD_SERVER)
+  file(GLOB onnxruntime_test_server_src
+    "${TEST_SRC_DIR}/server/unit_tests/*.cc"
+    "${TEST_SRC_DIR}/server/unit_tests/*.h"
   )
 
-  file(GLOB onnxruntime_integration_test_hosting_src
-    "${TEST_SRC_DIR}/hosting/integration_tests/*.py"
+  file(GLOB onnxruntime_integration_test_server_src
+    "${TEST_SRC_DIR}/server/integration_tests/*.py"
   )
   if(NOT WIN32)
     if(HAS_UNUSED_PARAMETER)
-      set_source_files_properties("${TEST_SRC_DIR}/hosting/unit_tests/json_handling_tests.cc" PROPERTIES COMPILE_FLAGS -Wno-unused-parameter)
-      set_source_files_properties("${TEST_SRC_DIR}/hosting/unit_tests/converter_tests.cc" PROPERTIES COMPILE_FLAGS -Wno-unused-parameter)
-      set_source_files_properties("${TEST_SRC_DIR}/hosting/unit_tests/util_tests.cc" PROPERTIES COMPILE_FLAGS -Wno-unused-parameter)
+      set_source_files_properties("${TEST_SRC_DIR}/server/unit_tests/json_handling_tests.cc" PROPERTIES COMPILE_FLAGS -Wno-unused-parameter)
+      set_source_files_properties("${TEST_SRC_DIR}/server/unit_tests/converter_tests.cc" PROPERTIES COMPILE_FLAGS -Wno-unused-parameter)
+      set_source_files_properties("${TEST_SRC_DIR}/server/unit_tests/util_tests.cc" PROPERTIES COMPILE_FLAGS -Wno-unused-parameter)
     endif()
   endif()
 
-  add_library(onnxruntime_test_utils_for_hosting ${onnxruntime_test_hosting_src})
-  onnxruntime_add_include_to_target(onnxruntime_test_utils_for_hosting onnxruntime_test_utils gtest gmock gsl onnx onnx_proto hosting_proto)
-  add_dependencies(onnxruntime_test_utils_for_hosting onnxruntime_hosting_lib onnxruntime_hosting_http_core_lib Boost ${onnxruntime_EXTERNAL_DEPENDENCIES})
-  target_include_directories(onnxruntime_test_utils_for_hosting PUBLIC ${Boost_INCLUDE_DIR} ${REPO_ROOT}/cmake/external/re2 ${CMAKE_CURRENT_BINARY_DIR}/onnx ${ONNXRUNTIME_ROOT}/hosting/http ${ONNXRUNTIME_ROOT}/hosting/http/core PRIVATE ${ONNXRUNTIME_ROOT} )
-  target_link_libraries(onnxruntime_test_utils_for_hosting ${Boost_LIBRARIES} ${onnx_test_libs})
+  add_library(onnxruntime_test_utils_for_server ${onnxruntime_test_server_src})
+  onnxruntime_add_include_to_target(onnxruntime_test_utils_for_server onnxruntime_test_utils gtest gmock gsl onnx onnx_proto server_proto)
+  add_dependencies(onnxruntime_test_utils_for_server onnxruntime_server_lib onnxruntime_server_http_core_lib Boost ${onnxruntime_EXTERNAL_DEPENDENCIES})
+  target_include_directories(onnxruntime_test_utils_for_server PUBLIC ${Boost_INCLUDE_DIR} ${REPO_ROOT}/cmake/external/re2 ${CMAKE_CURRENT_BINARY_DIR}/onnx ${ONNXRUNTIME_ROOT}/server/http ${ONNXRUNTIME_ROOT}/server/http/core PRIVATE ${ONNXRUNTIME_ROOT} )
+  target_link_libraries(onnxruntime_test_utils_for_server ${Boost_LIBRARIES} ${onnx_test_libs})
 
   AddTest(
-    TARGET onnxruntime_hosting_tests
-    SOURCES ${onnxruntime_test_hosting_src}
-    LIBS ${onnxruntime_test_hosting_libs} hosting_proto onnxruntime_hosting_lib ${onnxruntime_test_providers_libs}
+    TARGET onnxruntime_server_tests
+    SOURCES ${onnxruntime_test_server_src}
+    LIBS ${onnxruntime_test_server_libs} server_proto onnxruntime_server_lib ${onnxruntime_test_providers_libs}
     DEPENDS ${onnxruntime_EXTERNAL_DEPENDENCIES}
   )
 
   onnxruntime_protobuf_generate(
-          APPEND_PATH IMPORT_DIRS ${REPO_ROOT}/cmake/external/protobuf/src ${ONNXRUNTIME_ROOT}/hosting/protobuf ${ONNXRUNTIME_ROOT}/core/protobuf
-          PROTOS ${ONNXRUNTIME_ROOT}/hosting/protobuf/predict.proto ${ONNXRUNTIME_ROOT}/hosting/protobuf/onnx-ml.proto
+          APPEND_PATH IMPORT_DIRS ${REPO_ROOT}/cmake/external/protobuf/src ${ONNXRUNTIME_ROOT}/server/protobuf ${ONNXRUNTIME_ROOT}/core/protobuf
+          PROTOS ${ONNXRUNTIME_ROOT}/server/protobuf/predict.proto ${ONNXRUNTIME_ROOT}/server/protobuf/onnx-ml.proto
           LANGUAGE python
-          TARGET onnxruntime_hosting_tests
-          OUT_VAR hosting_test_py)
+          TARGET onnxruntime_server_tests
+          OUT_VAR server_test_py)
 
   add_custom_command(
-    TARGET onnxruntime_hosting_tests POST_BUILD
-    COMMAND ${CMAKE_COMMAND} -E make_directory $<TARGET_FILE_DIR:${test_data_target}>/hosting_test
+    TARGET onnxruntime_server_tests POST_BUILD
+    COMMAND ${CMAKE_COMMAND} -E make_directory ${CMAKE_CURRENT_BINARY_DIR}/server_test
     COMMAND ${CMAKE_COMMAND} -E copy
-      ${onnxruntime_integration_test_hosting_src}
-      $<TARGET_FILE_DIR:${test_data_target}>/hosting_test/
+      ${onnxruntime_integration_test_server_src}
+      ${CMAKE_CURRENT_BINARY_DIR}/server_test/
     COMMAND ${CMAKE_COMMAND} -E copy
-      $<TARGET_FILE_DIR:${test_data_target}>/*_pb2.py
-      $<TARGET_FILE_DIR:${test_data_target}>/hosting_test/
+      ${CMAKE_CURRENT_BINARY_DIR}/onnx_ml_pb2.py
+      ${CMAKE_CURRENT_BINARY_DIR}/server_test/
+    COMMAND ${CMAKE_COMMAND} -E copy
+      ${CMAKE_CURRENT_BINARY_DIR}/predict_pb2.py
+      ${CMAKE_CURRENT_BINARY_DIR}/server_test/
   )
 
 endif()
diff --git a/docs/Hosting_Application_Usage.md b/docs/ONNX_Runtime_Server_Usage.md
similarity index 88%
rename from docs/Hosting_Application_Usage.md
rename to docs/ONNX_Runtime_Server_Usage.md
index a542ea32a46a9..ac23bbb83726a 100644
--- a/docs/Hosting_Application_Usage.md
+++ b/docs/ONNX_Runtime_Server_Usage.md
@@ -1,21 +1,21 @@
-<h1><span style="color:red">Note: Hosting Application is still in beta state. It's currently not ready for production environments.</span></h1>
+<h1><span style="color:red">Note: ONNX Runtime Server is still in beta state. It's currently not ready for production environments.</span></h1>
 
-# How to Use ONNX Hosting Application REST API for Prediction
+# How to Use ONNX Runtime Server REST API for Prediction
 
-ONNX Hosting provides a REST API for prediction. The goal of the project is to make it easy to "host" any ONNX model as a RESTful service. The CLI command to start the service is shown below:
+ONNX Runtime Server provides a REST API for prediction. The goal of the project is to make it easy to "host" any ONNX model as a RESTful service. The CLI command to start the service is shown below:
 
 ```
-~./onnxruntime_hosting --help
-ONNX Hosting: host an ONNX model for inferencing with ONNXRuntime
+$ ./onnxruntime_server
+the option '--model_path' is required but missing
 Allowed options:
-  -h [ --help ]                   Shows a help message and exits
-  --log_level arg (=info)         Logging level. Allowed options (case 
-                                  sensitive): verbose, info, warning, error, 
-                                  fatal
-  -m [ --model_path ] arg         Path to ONNX model
-  -a [ --address ] arg (=0.0.0.0) The base HTTP address
-  --http_port arg (=8001)         HTTP port to listen to requests
-  --num_http_threads arg (=8)     Number of http threads
+  -h [ --help ]                Shows a help message and exits
+  --log_level arg (=info)      Logging level. Allowed options (case sensitive):
+                               verbose, info, warning, error, fatal
+  --model_path arg             Path to ONNX model
+  --address arg (=0.0.0.0)     The base HTTP address
+  --http_port arg (=8001)      HTTP port to listen to requests
+  --num_http_threads arg (=<# of your cpu cores>) Number of http threads
+
 
 ```
 
@@ -26,7 +26,7 @@ Note: The only mandatory argument for the program here is `model_path`
 To host an ONNX model as a REST API server, run:
 
 ```
-./onnxruntime_hosting -m /<your>/<model>/<path>
+./onnxruntime_server --model_path /<your>/<model>/<path>
 ```
 
 The prediction URL is in this format:
@@ -44,7 +44,7 @@ An HTTP request can be a Protobuf message in two formats: binary or JSON. The HT
 * For `"Content-Type: application/json"`, the payload will be deserialized as JSON string in UTF-8 format
 * For `"Content-Type: application/vnd.google.protobuf"`, `"Content-Type: application/x-protobuf"` or `"Content-Type: application/octet-stream"`, the payload will be consumed as protobuf message directly.
 
-The Protobuf definition can be found [here](https://github.com/Microsoft/onnxruntime/blob/master/onnxruntime/hosting/protobuf/predict.proto).
+The Protobuf definition can be found [here](https://github.com/Microsoft/onnxruntime/blob/master/onnxruntime/server/protobuf/predict.proto).
 
 ## Inferencing
 
diff --git a/onnxruntime/hosting/converter.cc b/onnxruntime/server/converter.cc
similarity index 99%
rename from onnxruntime/hosting/converter.cc
rename to onnxruntime/server/converter.cc
index 5b2852fe7bc1b..56fd51cd10c4c 100644
--- a/onnxruntime/hosting/converter.cc
+++ b/onnxruntime/server/converter.cc
@@ -17,7 +17,7 @@
 #include "converter.h"
 
 namespace onnxruntime {
-namespace hosting {
+namespace server {
 
 namespace protobufutil = google::protobuf::util;
 
@@ -257,5 +257,5 @@ common::Status MLValueToTensorProto(const onnxruntime::MLValue& ml_value, bool u
 
   return common::Status::OK();
 }
-}  // namespace hosting
+}  // namespace server
 }  // namespace onnxruntime
\ No newline at end of file
diff --git a/onnxruntime/hosting/converter.h b/onnxruntime/server/converter.h
similarity index 96%
rename from onnxruntime/hosting/converter.h
rename to onnxruntime/server/converter.h
index bfa64cf9ca009..983c9f454596c 100644
--- a/onnxruntime/hosting/converter.h
+++ b/onnxruntime/server/converter.h
@@ -11,7 +11,7 @@
 #include "predict.pb.h"
 
 namespace onnxruntime {
-namespace hosting {
+namespace server {
 
 onnx::TensorProto_DataType MLDataTypeToTensorProtoDataType(const onnxruntime::DataTypeImpl* cpp_type);
 
@@ -25,5 +25,5 @@ common::Status MLValueToTensorProto(const onnxruntime::MLValue& ml_value, bool u
                                     std::unique_ptr<onnxruntime::logging::Logger> logger,
                                     /* out */ onnx::TensorProto& tensor_proto);
 
-}  // namespace hosting
+}  // namespace server
 }  // namespace onnxruntime
diff --git a/onnxruntime/hosting/environment.cc b/onnxruntime/server/environment.cc
similarity index 74%
rename from onnxruntime/hosting/environment.cc
rename to onnxruntime/server/environment.cc
index 51352b9f99346..5235cad266d8c 100644
--- a/onnxruntime/hosting/environment.cc
+++ b/onnxruntime/server/environment.cc
@@ -8,10 +8,10 @@
 #include "log_sink.h"
 
 namespace onnxruntime {
-namespace hosting {
+namespace server {
 
-HostingEnvironment::HostingEnvironment(logging::Severity severity) : severity_(severity),
-                                                                     logger_id_("HostingApp"),
+ServerEnvironment::ServerEnvironment(logging::Severity severity) : severity_(severity),
+                                                                     logger_id_("ServerApp"),
                                                                      default_logging_manager_(
                                                                          std::unique_ptr<logging::ISink>{new LogSink{}},
                                                                          severity,
@@ -24,7 +24,7 @@ HostingEnvironment::HostingEnvironment(logging::Severity severity) : severity_(s
   session = std::make_unique<onnxruntime::InferenceSession>(options_, &default_logging_manager_);
 }
 
-common::Status HostingEnvironment::InitializeModel(const std::string& model_path) {
+common::Status ServerEnvironment::InitializeModel(const std::string& model_path) {
   auto status = session->Load(model_path);
   if (!status.IsOK()) {
     return status;
@@ -42,19 +42,19 @@ common::Status HostingEnvironment::InitializeModel(const std::string& model_path
   return common::Status::OK();
 }
 
-const std::vector<std::string>& HostingEnvironment::GetModelOutputNames() const {
+const std::vector<std::string>& ServerEnvironment::GetModelOutputNames() const {
   return model_output_names_;
 }
 
-const logging::Logger& HostingEnvironment::GetAppLogger() const {
+const logging::Logger& ServerEnvironment::GetAppLogger() const {
   return default_logging_manager_.DefaultLogger();
 }
 
-logging::Severity HostingEnvironment::GetLogSeverity() const {
+logging::Severity ServerEnvironment::GetLogSeverity() const {
   return severity_;
 }
 
-std::unique_ptr<logging::Logger> HostingEnvironment::GetLogger(const std::string& id) {
+std::unique_ptr<logging::Logger> ServerEnvironment::GetLogger(const std::string& id) {
   if (id.empty()) {
     LOGS(GetAppLogger(), WARNING) << "Request id is null or empty string";
   }
@@ -62,9 +62,9 @@ std::unique_ptr<logging::Logger> HostingEnvironment::GetLogger(const std::string
   return default_logging_manager_.CreateLogger(id, severity_, false);
 }
 
-onnxruntime::InferenceSession* HostingEnvironment::GetSession() const {
+onnxruntime::InferenceSession* ServerEnvironment::GetSession() const {
   return session.get();
 }
 
-}  // namespace hosting
+}  // namespace server
 }  // namespace onnxruntime
diff --git a/onnxruntime/hosting/environment.h b/onnxruntime/server/environment.h
similarity index 82%
rename from onnxruntime/hosting/environment.h
rename to onnxruntime/server/environment.h
index 011fc1336710f..4e0e408a7f20f 100644
--- a/onnxruntime/hosting/environment.h
+++ b/onnxruntime/server/environment.h
@@ -11,15 +11,15 @@
 #include "core/session/inference_session.h"
 
 namespace onnxruntime {
-namespace hosting {
+namespace server {
 
 namespace logging = logging;
 
-class HostingEnvironment {
+class ServerEnvironment {
  public:
-  explicit HostingEnvironment(logging::Severity severity);
-  ~HostingEnvironment() = default;
-  HostingEnvironment(const HostingEnvironment&) = delete;
+  explicit ServerEnvironment(logging::Severity severity);
+  ~ServerEnvironment() = default;
+  ServerEnvironment(const ServerEnvironment&) = delete;
 
   const logging::Logger& GetAppLogger() const;
   std::unique_ptr<logging::Logger> GetLogger(const std::string& id);
@@ -41,5 +41,5 @@ class HostingEnvironment {
   std::vector<std::string> model_output_names_;
 };
 
-}  // namespace hosting
+}  // namespace server
 }  // namespace onnxruntime
diff --git a/onnxruntime/hosting/executor.cc b/onnxruntime/server/executor.cc
similarity index 94%
rename from onnxruntime/hosting/executor.cc
rename to onnxruntime/server/executor.cc
index 0a3f26d215548..48535129b9028 100644
--- a/onnxruntime/hosting/executor.cc
+++ b/onnxruntime/server/executor.cc
@@ -20,7 +20,7 @@
 #include "util.h"
 
 namespace onnxruntime {
-namespace hosting {
+namespace server {
 
 namespace protobufutil = google::protobuf::util;
 
@@ -51,7 +51,7 @@ protobufutil::Status Executor::SetMLValue(const onnx::TensorProto& input_tensor,
   return protobufutil::Status::OK;
 }
 
-protobufutil::Status Executor::SetNameMLValueMap(onnxruntime::NameMLValMap& name_value_map, const onnxruntime::hosting::PredictRequest& request) {
+protobufutil::Status Executor::SetNameMLValueMap(onnxruntime::NameMLValMap& name_value_map, const onnxruntime::server::PredictRequest& request) {
   auto logger = env_->GetLogger(request_id_);
 
   OrtAllocatorInfo* cpu_allocator_info = nullptr;
@@ -84,8 +84,8 @@ protobufutil::Status Executor::SetNameMLValueMap(onnxruntime::NameMLValMap& name
 
 protobufutil::Status Executor::Predict(const std::string& model_name,
                                        const std::string& model_version,
-                                       onnxruntime::hosting::PredictRequest& request,
-                                       /* out */ onnxruntime::hosting::PredictResponse& response) {
+                                       onnxruntime::server::PredictRequest& request,
+                                       /* out */ onnxruntime::server::PredictResponse& response) {
   auto logger = env_->GetLogger(request_id_);
 
   // Convert PredictRequest to NameMLValMap
@@ -144,5 +144,5 @@ protobufutil::Status Executor::Predict(const std::string& model_name,
   return protobufutil::Status::OK;
 }
 
-}  // namespace hosting
+}  // namespace server
 }  // namespace onnxruntime
\ No newline at end of file
diff --git a/onnxruntime/hosting/executor.h b/onnxruntime/server/executor.h
similarity index 75%
rename from onnxruntime/hosting/executor.h
rename to onnxruntime/server/executor.h
index 17a0fb5ab8321..47aaa49b84d07 100644
--- a/onnxruntime/hosting/executor.h
+++ b/onnxruntime/server/executor.h
@@ -9,22 +9,22 @@
 #include "predict.pb.h"
 
 namespace onnxruntime {
-namespace hosting {
+namespace server {
 
 class Executor {
  public:
-  Executor(HostingEnvironment* hosting_env, std::string request_id) : env_(hosting_env),
+  Executor(ServerEnvironment* server_env, std::string request_id) : env_(server_env),
                                                                                       request_id_(std::move(request_id)),
                                                                                       using_raw_data_(true) {}
 
   // Prediction method
   google::protobuf::util::Status Predict(const std::string& model_name,
                                          const std::string& model_version,
-                                         onnxruntime::hosting::PredictRequest& request,
-                                         /* out */ onnxruntime::hosting::PredictResponse& response);
+                                         onnxruntime::server::PredictRequest& request,
+                                         /* out */ onnxruntime::server::PredictResponse& response);
 
  private:
-  HostingEnvironment* env_;
+  ServerEnvironment* env_;
   const std::string request_id_;
   bool using_raw_data_;
 
@@ -32,8 +32,8 @@ class Executor {
                                             OrtAllocatorInfo* cpu_allocator_info,
                                             /* out */ MLValue& ml_value);
 
-  google::protobuf::util::Status SetNameMLValueMap(onnxruntime::NameMLValMap& name_value_map, const onnxruntime::hosting::PredictRequest& request);
+  google::protobuf::util::Status SetNameMLValueMap(onnxruntime::NameMLValMap& name_value_map, const onnxruntime::server::PredictRequest& request);
 };
 
-}  // namespace hosting
+}  // namespace server
 }  // namespace onnxruntime
diff --git a/onnxruntime/hosting/http/core/context.h b/onnxruntime/server/http/core/context.h
similarity index 96%
rename from onnxruntime/hosting/http/core/context.h
rename to onnxruntime/server/http/core/context.h
index 63ad0ef009a10..b9598762fd5a5 100644
--- a/onnxruntime/hosting/http/core/context.h
+++ b/onnxruntime/server/http/core/context.h
@@ -16,7 +16,7 @@
 #include <boost/uuid/uuid_generators.hpp>
 
 namespace onnxruntime {
-namespace hosting {
+namespace server {
 
 namespace http = boost::beast::http;  // from <boost/beast/http.hpp>
 
@@ -42,5 +42,5 @@ class HttpContext {
   HttpContext(const HttpContext&) = delete;
 };
 
-}  // namespace hosting
+}  // namespace server
 }  // namespace onnxruntime
diff --git a/onnxruntime/hosting/http/core/http_server.cc b/onnxruntime/server/http/core/http_server.cc
similarity index 97%
rename from onnxruntime/hosting/http/core/http_server.cc
rename to onnxruntime/server/http/core/http_server.cc
index 134e8a09d83a9..a7ae6f7cbe6c1 100644
--- a/onnxruntime/hosting/http/core/http_server.cc
+++ b/onnxruntime/server/http/core/http_server.cc
@@ -20,7 +20,7 @@ namespace net = boost::asio;          // from <boost/asio.hpp>
 using tcp = boost::asio::ip::tcp;     // from <boost/asio/ip/tcp.hpp>
 
 namespace onnxruntime {
-namespace hosting {
+namespace server {
 
 App::App() {
   http_details.address = boost::asio::ip::make_address_v4("0.0.0.0");
@@ -84,5 +84,5 @@ App& App::Run() {
   ioc.run();
   return *this;
 }
-}  // namespace hosting
+}  // namespace server
 }  // namespace onnxruntime
diff --git a/onnxruntime/hosting/http/core/http_server.h b/onnxruntime/server/http/core/http_server.h
similarity index 96%
rename from onnxruntime/hosting/http/core/http_server.h
rename to onnxruntime/server/http/core/http_server.h
index e28d12939589c..505d93668cb47 100644
--- a/onnxruntime/hosting/http/core/http_server.h
+++ b/onnxruntime/server/http/core/http_server.h
@@ -17,7 +17,7 @@
 #include "listener.h"
 
 namespace onnxruntime {
-namespace hosting {
+namespace server {
 
 namespace http = beast::http;      // from <boost/beast/http.hpp>
 namespace net = boost::asio;       // from <boost/asio.hpp>
@@ -49,5 +49,5 @@ class App {
   StartFn on_start_ = {};
   Details http_details{};
 };
-}  // namespace hosting
+}  // namespace server
 }  // namespace onnxruntime
diff --git a/onnxruntime/hosting/http/core/listener.cc b/onnxruntime/server/http/core/listener.cc
similarity index 97%
rename from onnxruntime/hosting/http/core/listener.cc
rename to onnxruntime/server/http/core/listener.cc
index c30b936119df2..93e2e4ac68eae 100644
--- a/onnxruntime/hosting/http/core/listener.cc
+++ b/onnxruntime/server/http/core/listener.cc
@@ -6,7 +6,7 @@
 #include "util.h"
 
 namespace onnxruntime {
-namespace hosting {
+namespace server {
 
 namespace net = boost::asio;       // from <boost/asio.hpp>
 using tcp = boost::asio::ip::tcp;  // from <boost/asio/ip/tcp.hpp>
@@ -78,5 +78,5 @@ void Listener::OnAccept(beast::error_code ec) {
   // Accept another connection
   DoAccept();
 }
-}  // namespace hosting
+}  // namespace server
 }  // namespace onnxruntime
\ No newline at end of file
diff --git a/onnxruntime/hosting/http/core/listener.h b/onnxruntime/server/http/core/listener.h
similarity index 95%
rename from onnxruntime/hosting/http/core/listener.h
rename to onnxruntime/server/http/core/listener.h
index 0415c0fb93c27..3295e6a448cb4 100644
--- a/onnxruntime/hosting/http/core/listener.h
+++ b/onnxruntime/server/http/core/listener.h
@@ -11,7 +11,7 @@
 #include "util.h"
 
 namespace onnxruntime {
-namespace hosting {
+namespace server {
 
 namespace net = boost::asio;       // from <boost/asio.hpp>
 using tcp = boost::asio::ip::tcp;  // from <boost/asio/ip/tcp.hpp>
@@ -39,6 +39,6 @@ class Listener : public std::enable_shared_from_this<Listener> {
   void OnAccept(beast::error_code ec);
 };
 
-}  // namespace hosting
+}  // namespace server
 }  // namespace onnxruntime
 
diff --git a/onnxruntime/hosting/http/core/routes.cc b/onnxruntime/server/http/core/routes.cc
similarity index 97%
rename from onnxruntime/hosting/http/core/routes.cc
rename to onnxruntime/server/http/core/routes.cc
index af174c32096a3..5bb0878845b34 100644
--- a/onnxruntime/hosting/http/core/routes.cc
+++ b/onnxruntime/server/http/core/routes.cc
@@ -8,7 +8,7 @@
 #include "routes.h"
 
 namespace onnxruntime {
-namespace hosting {
+namespace server {
 
 namespace http = boost::beast::http;  // from <boost/beast/http.hpp>
 
@@ -77,5 +77,5 @@ http::status Routes::ParseUrl(http::verb method,
   return http::status::ok;
 }
 
-}  //namespace hosting
+}  //namespace server
 }  // namespace onnxruntime
\ No newline at end of file
diff --git a/onnxruntime/hosting/http/core/routes.h b/onnxruntime/server/http/core/routes.h
similarity index 96%
rename from onnxruntime/hosting/http/core/routes.h
rename to onnxruntime/server/http/core/routes.h
index 29a1064edf7b0..5681f2437d605 100644
--- a/onnxruntime/hosting/http/core/routes.h
+++ b/onnxruntime/server/http/core/routes.h
@@ -8,7 +8,7 @@
 #include "context.h"
 
 namespace onnxruntime {
-namespace hosting {
+namespace server {
 
 namespace http = boost::beast::http;  // from <boost/beast/http.hpp>
 
@@ -36,6 +36,6 @@ class Routes {
   std::vector<std::pair<std::string, HandlerFn>> get_fn_table;
 };
 
-}  //namespace hosting
+}  //namespace server
 }  // namespace onnxruntime
 
diff --git a/onnxruntime/hosting/http/core/session.cc b/onnxruntime/server/http/core/session.cc
similarity index 99%
rename from onnxruntime/hosting/http/core/session.cc
rename to onnxruntime/server/http/core/session.cc
index 6fe9eec742d6f..e7463ade65226 100644
--- a/onnxruntime/hosting/http/core/session.cc
+++ b/onnxruntime/server/http/core/session.cc
@@ -4,7 +4,7 @@
 #include "session.h"
 
 namespace onnxruntime {
-namespace hosting {
+namespace server {
 
 namespace net = boost::asio;       // from <boost/asio.hpp>
 namespace beast = boost::beast;    // from <boost/beast.hpp>
@@ -148,5 +148,5 @@ http::status HttpSession::ExecuteUserFunction(HttpContext& context) {
   return http::status::ok;
 }
 
-}  // namespace hosting
+}  // namespace server
 }  // namespace onnxruntime
\ No newline at end of file
diff --git a/onnxruntime/hosting/http/core/session.h b/onnxruntime/server/http/core/session.h
similarity index 98%
rename from onnxruntime/hosting/http/core/session.h
rename to onnxruntime/server/http/core/session.h
index a3827c52ba9c4..c1c8dd5f3c0a6 100644
--- a/onnxruntime/hosting/http/core/session.h
+++ b/onnxruntime/server/http/core/session.h
@@ -15,7 +15,7 @@
 #include "util.h"
 
 namespace onnxruntime {
-namespace hosting {
+namespace server {
 
 namespace net = boost::asio;       // from <boost/asio.hpp>
 namespace beast = boost::beast;    // from <boost/beast.hpp>
@@ -74,6 +74,6 @@ class HttpSession : public std::enable_shared_from_this<HttpSession> {
   void DoClose();
 };
 
-}  // namespace hosting
+}  // namespace server
 }  // namespace onnxruntime
 
diff --git a/onnxruntime/hosting/http/core/util.cc b/onnxruntime/server/http/core/util.cc
similarity index 90%
rename from onnxruntime/hosting/http/core/util.cc
rename to onnxruntime/server/http/core/util.cc
index eb23a067bac76..a5680b800c4e4 100644
--- a/onnxruntime/hosting/http/core/util.cc
+++ b/onnxruntime/server/http/core/util.cc
@@ -9,12 +9,12 @@
 #include "util.h"
 
 namespace onnxruntime {
-namespace hosting {
+namespace server {
 
 // Report a failure
 void ErrorHandling(beast::error_code ec, char const* what) {
   std::cerr << what << " failed: " << ec.value() << " : " << ec.message() << "\n";
 }
 
-}  // namespace hosting
+}  // namespace server
 }  // namespace onnxruntime
\ No newline at end of file
diff --git a/onnxruntime/hosting/http/core/util.h b/onnxruntime/server/http/core/util.h
similarity index 90%
rename from onnxruntime/hosting/http/core/util.h
rename to onnxruntime/server/http/core/util.h
index 5d73000c2a866..54faea9629ff8 100644
--- a/onnxruntime/hosting/http/core/util.h
+++ b/onnxruntime/server/http/core/util.h
@@ -9,13 +9,13 @@
 #include "context.h"
 
 namespace onnxruntime {
-namespace hosting {
+namespace server {
 
 namespace beast = boost::beast;  // from <boost/beast.hpp>
 
 // Report a failure
 void ErrorHandling(beast::error_code ec, char const* what);
 
-}  // namespace hosting
+}  // namespace server
 }  // namespace onnxruntime
 
diff --git a/onnxruntime/hosting/http/json_handling.cc b/onnxruntime/server/http/json_handling.cc
similarity index 88%
rename from onnxruntime/hosting/http/json_handling.cc
rename to onnxruntime/server/http/json_handling.cc
index caae91b00a5b6..6bc46b7878338 100644
--- a/onnxruntime/hosting/http/json_handling.cc
+++ b/onnxruntime/server/http/json_handling.cc
@@ -13,9 +13,9 @@
 namespace protobufutil = google::protobuf::util;
 
 namespace onnxruntime {
-namespace hosting {
+namespace server {
 
-protobufutil::Status GetRequestFromJson(const std::string& json_string, /* out */ onnxruntime::hosting::PredictRequest& request) {
+protobufutil::Status GetRequestFromJson(const std::string& json_string, /* out */ onnxruntime::server::PredictRequest& request) {
   protobufutil::JsonParseOptions options;
   options.ignore_unknown_fields = true;
 
@@ -23,7 +23,7 @@ protobufutil::Status GetRequestFromJson(const std::string& json_string, /* out *
   return result;
 }
 
-protobufutil::Status GenerateResponseInJson(const onnxruntime::hosting::PredictResponse& response, /* out */ std::string& json_string) {
+protobufutil::Status GenerateResponseInJson(const onnxruntime::server::PredictResponse& response, /* out */ std::string& json_string) {
   protobufutil::JsonPrintOptions options;
   options.add_whitespace = false;
   options.always_print_primitive_fields = false;
@@ -62,5 +62,5 @@ std::string escape_string(const std::string& message) {
   return o.str();
 }
 
-}  // namespace hosting
+}  // namespace server
 }  // namespace onnxruntime
\ No newline at end of file
diff --git a/onnxruntime/hosting/http/json_handling.h b/onnxruntime/server/http/json_handling.h
similarity index 85%
rename from onnxruntime/hosting/http/json_handling.h
rename to onnxruntime/server/http/json_handling.h
index ce6c7f9e9cbeb..1e3d8f7239db1 100644
--- a/onnxruntime/hosting/http/json_handling.h
+++ b/onnxruntime/server/http/json_handling.h
@@ -9,18 +9,18 @@
 #include "predict.pb.h"
 
 namespace onnxruntime {
-namespace hosting {
+namespace server {
 
 namespace http = boost::beast::http;
 
 // Deserialize Json input to PredictRequest.
 // Unknown fields in the json file will be ignored.
-google::protobuf::util::Status GetRequestFromJson(const std::string& json_string, /* out */ onnxruntime::hosting::PredictRequest& request);
+google::protobuf::util::Status GetRequestFromJson(const std::string& json_string, /* out */ onnxruntime::server::PredictRequest& request);
 
 // Serialize PredictResponse to json string
 // 1. Proto3 primitive fields with default values will be omitted in JSON output. Eg. int32 field with value 0 will be omitted
 // 2. Enums will be printed as string, not int, to improve readability
-google::protobuf::util::Status GenerateResponseInJson(const onnxruntime::hosting::PredictResponse& response, /* out */ std::string& json_string);
+google::protobuf::util::Status GenerateResponseInJson(const onnxruntime::server::PredictResponse& response, /* out */ std::string& json_string);
 
 // Constructs JSON error message from error code object and error message
 std::string CreateJsonError(http::status error_code, const std::string& error_message);
@@ -29,6 +29,6 @@ std::string CreateJsonError(http::status error_code, const std::string& error_me
 // Mostly taken from here: https://stackoverflow.com/questions/7724448/simple-json-string-escape-for-c/33799784#33799784
 std::string escape_string(const std::string& message);
 
-}  // namespace hosting
+}  // namespace server
 }  // namespace onnxruntime
 
diff --git a/onnxruntime/hosting/http/predict_request_handler.cc b/onnxruntime/server/http/predict_request_handler.cc
similarity index 98%
rename from onnxruntime/hosting/http/predict_request_handler.cc
rename to onnxruntime/server/http/predict_request_handler.cc
index f3fadd32c53db..32d6443744e3d 100644
--- a/onnxruntime/hosting/http/predict_request_handler.cc
+++ b/onnxruntime/server/http/predict_request_handler.cc
@@ -10,7 +10,7 @@
 #include "util.h"
 
 namespace onnxruntime {
-namespace hosting {
+namespace server {
 
 namespace protobufutil = google::protobuf::util;
 
@@ -35,7 +35,7 @@ void Predict(const std::string& name,
              const std::string& version,
              const std::string& action,
              /* in, out */ HttpContext& context,
-             std::shared_ptr<HostingEnvironment> env) {
+             std::shared_ptr<ServerEnvironment> env) {
   auto logger = env->GetLogger(context.request_id);
   LOGS(*logger, INFO) << "Model Name: " << name << ", Version: " << version << ", Action: " << action;
 
@@ -130,5 +130,5 @@ static bool ParseRequestPayload(const HttpContext& context, SupportedContentType
   return true;
 }
 
-}  // namespace hosting
+}  // namespace server
 }  // namespace onnxruntime
diff --git a/onnxruntime/hosting/http/predict_request_handler.h b/onnxruntime/server/http/predict_request_handler.h
similarity index 84%
rename from onnxruntime/hosting/http/predict_request_handler.h
rename to onnxruntime/server/http/predict_request_handler.h
index 0b1cce1216f36..8fe0d7f74d263 100644
--- a/onnxruntime/hosting/http/predict_request_handler.h
+++ b/onnxruntime/server/http/predict_request_handler.h
@@ -5,7 +5,7 @@
 #include "json_handling.h"
 
 namespace onnxruntime {
-namespace hosting {
+namespace server {
 
 namespace beast = boost::beast;
 namespace http = beast::http;
@@ -17,7 +17,7 @@ void Predict(const std::string& name,
              const std::string& version,
              const std::string& action,
              /* in, out */ HttpContext& context,
-             std::shared_ptr<HostingEnvironment> env);
+             std::shared_ptr<ServerEnvironment> env);
 
-}  // namespace hosting
+}  // namespace server
 }  // namespace onnxruntime
diff --git a/onnxruntime/hosting/http/util.cc b/onnxruntime/server/http/util.cc
similarity index 98%
rename from onnxruntime/hosting/http/util.cc
rename to onnxruntime/server/http/util.cc
index 0b9d2a704e7e0..61986291f444a 100644
--- a/onnxruntime/hosting/http/util.cc
+++ b/onnxruntime/server/http/util.cc
@@ -11,7 +11,7 @@
 
 namespace protobufutil = google::protobuf::util;
 namespace onnxruntime {
-namespace hosting {
+namespace server {
 
 static std::unordered_set<std::string> protobuf_mime_types{
     "application/octet-stream",
@@ -80,5 +80,5 @@ SupportedContentType GetResponseContentType(const HttpContext& context) {
   return SupportedContentType::Unknown;
 }
 
-}  // namespace hosting
+}  // namespace server
 }  // namespace onnxruntime
\ No newline at end of file
diff --git a/onnxruntime/hosting/http/util.h b/onnxruntime/server/http/util.h
similarity index 92%
rename from onnxruntime/hosting/http/util.h
rename to onnxruntime/server/http/util.h
index 469d40e4bda1a..ba38b3976a1cb 100644
--- a/onnxruntime/hosting/http/util.h
+++ b/onnxruntime/server/http/util.h
@@ -7,10 +7,10 @@
 #include <boost/beast/http/status.hpp>
 #include <google/protobuf/stubs/status.h>
 
-#include "hosting/http/core/context.h"
+#include "server/http/core/context.h"
 
 namespace onnxruntime {
-namespace hosting {
+namespace server {
 
 namespace beast = boost::beast;  // from <boost/beast.hpp>
 
@@ -31,5 +31,5 @@ SupportedContentType GetRequestContentType(const HttpContext& context);
 // Currently we only support three types of response content type: */*, application/json and application/octet-stream
 SupportedContentType GetResponseContentType(const HttpContext& context);
 
-}  // namespace hosting
+}  // namespace server
 }  // namespace onnxruntime
diff --git a/onnxruntime/hosting/log_sink.h b/onnxruntime/server/log_sink.h
similarity index 90%
rename from onnxruntime/hosting/log_sink.h
rename to onnxruntime/server/log_sink.h
index b177fec6052fc..70df47f72c0a7 100644
--- a/onnxruntime/hosting/log_sink.h
+++ b/onnxruntime/server/log_sink.h
@@ -8,13 +8,13 @@
 #include "core/common/logging/sinks/ostream_sink.h"
 
 namespace onnxruntime {
-namespace hosting {
+namespace server {
 
 class LogSink : public onnxruntime::logging::OStreamSink {
  public:
   LogSink() : OStreamSink(std::cout, /*flush*/ true) {
   }
 };
-}  // namespace hosting
+}  // namespace server
 }  // namespace onnxruntime
 
diff --git a/onnxruntime/hosting/main.cc b/onnxruntime/server/main.cc
similarity index 84%
rename from onnxruntime/hosting/main.cc
rename to onnxruntime/server/main.cc
index 88a32942852e8..ac4891566b1d5 100644
--- a/onnxruntime/hosting/main.cc
+++ b/onnxruntime/server/main.cc
@@ -8,19 +8,19 @@
 
 namespace beast = boost::beast;
 namespace http = beast::http;
-namespace hosting = onnxruntime::hosting;
+namespace server = onnxruntime::server;
 
 int main(int argc, char* argv[]) {
-  hosting::ServerConfiguration config{};
+  server::ServerConfiguration config{};
   auto res = config.ParseInput(argc, argv);
 
-  if (res == hosting::Result::ExitSuccess) {
+  if (res == server::Result::ExitSuccess) {
     exit(EXIT_SUCCESS);
-  } else if (res == hosting::Result::ExitFailure) {
+  } else if (res == server::Result::ExitFailure) {
     exit(EXIT_FAILURE);
   }
 
-  auto env = std::make_shared<hosting::HostingEnvironment>(config.logging_level);
+  auto env = std::make_shared<server::ServerEnvironment>(config.logging_level);
   auto logger = env->GetAppLogger();
   LOGS(logger, VERBOSE) << "Logging manager initialized.";
   LOGS(logger, INFO) << "Model path: " << config.model_path;
@@ -42,7 +42,7 @@ int main(int argc, char* argv[]) {
   }
 
   auto const boost_address = boost::asio::ip::make_address(config.address);
-  hosting::App app{};
+  server::App app{};
 
   app.RegisterStartup(
       [env](const auto& details) -> void {
@@ -63,13 +63,13 @@ int main(int argc, char* argv[]) {
         if (!context.client_request_id.empty()) {
           context.response.insert("x-ms-client-request-id", (context).client_request_id);
         }
-        context.response.body() = hosting::CreateJsonError(context.error_code, context.error_message);
+        context.response.body() = server::CreateJsonError(context.error_code, context.error_message);
       });
 
   app.RegisterPost(
       R"(/v1/models/([^/:]+)(?:/versions/(\d+))?:(classify|regress|predict))",
       [env](const auto& name, const auto& version, const auto& action, auto& context) -> void {
-        hosting::Predict(name, version, action, context, env);
+        server::Predict(name, version, action, context, env);
       });
 
   app.Bind(boost_address, config.http_port)
diff --git a/onnxruntime/hosting/protobuf/onnx-ml.proto b/onnxruntime/server/protobuf/onnx-ml.proto
similarity index 100%
rename from onnxruntime/hosting/protobuf/onnx-ml.proto
rename to onnxruntime/server/protobuf/onnx-ml.proto
diff --git a/onnxruntime/hosting/protobuf/predict.proto b/onnxruntime/server/protobuf/predict.proto
similarity index 96%
rename from onnxruntime/hosting/protobuf/predict.proto
rename to onnxruntime/server/protobuf/predict.proto
index 33029ad8a600c..21b04386353eb 100644
--- a/onnxruntime/hosting/protobuf/predict.proto
+++ b/onnxruntime/server/protobuf/predict.proto
@@ -2,7 +2,7 @@ syntax = "proto3";
 
 import "onnx-ml.proto";
 
-package onnxruntime.hosting;
+package onnxruntime.server;
 
 // PredictRequest specifies how inputs are mapped to tensors  
 // and how outputs are filtered before returning to user. 
diff --git a/onnxruntime/hosting/server_configuration.h b/onnxruntime/server/server_configuration.h
similarity index 97%
rename from onnxruntime/hosting/server_configuration.h
rename to onnxruntime/server/server_configuration.h
index d5b1ee459bc21..4ae9b58496da7 100644
--- a/onnxruntime/hosting/server_configuration.h
+++ b/onnxruntime/server/server_configuration.h
@@ -11,7 +11,7 @@
 #include "core/common/logging/logging.h"
 
 namespace onnxruntime {
-namespace hosting {
+namespace server {
 
 namespace po = boost::program_options;
 
@@ -37,7 +37,7 @@ static std::unordered_map<std::string, onnxruntime::logging::Severity> supported
 // Provides sane default values
 class ServerConfiguration {
  public:
-  const std::string full_desc = "ONNX Hosting: host an ONNX model with ONNX Runtime";
+  const std::string full_desc = "ONNX Server: host an ONNX model with ONNX Runtime";
   std::string model_path;
   std::string address = "0.0.0.0";
   int http_port = 8001;
@@ -127,5 +127,5 @@ class ServerConfiguration {
   }
 };
 
-}  // namespace hosting
+}  // namespace server
 }  // namespace onnxruntime
diff --git a/onnxruntime/hosting/util.cc b/onnxruntime/server/util.cc
similarity index 97%
rename from onnxruntime/hosting/util.cc
rename to onnxruntime/server/util.cc
index 47bc4f03cb183..579c2e81bcc3c 100644
--- a/onnxruntime/hosting/util.cc
+++ b/onnxruntime/server/util.cc
@@ -8,7 +8,7 @@
 #include "util.h"
 
 namespace onnxruntime {
-namespace hosting {
+namespace server {
 
 namespace protobufutil = google::protobuf::util;
 
@@ -44,5 +44,5 @@ protobufutil::Status GenerateProtobufStatus(const onnxruntime::common::Status& o
   return protobufutil::Status(code, oss.str());
 }
 
-}  // namespace hosting
+}  // namespace server
 }  // namespace onnxruntime
diff --git a/onnxruntime/hosting/util.h b/onnxruntime/server/util.h
similarity index 90%
rename from onnxruntime/hosting/util.h
rename to onnxruntime/server/util.h
index ead9e866d6a8f..e46a3a13fac63 100644
--- a/onnxruntime/hosting/util.h
+++ b/onnxruntime/server/util.h
@@ -8,11 +8,11 @@
 #include "core/common/status.h"
 
 namespace onnxruntime {
-namespace hosting {
+namespace server {
 
 // Generate protobuf status from ONNX Runtime status
 google::protobuf::util::Status GenerateProtobufStatus(const onnxruntime::common::Status& onnx_status, const std::string& message);
 
-}  // namespace hosting
+}  // namespace server
 }  // namespace onnxruntime
 
diff --git a/onnxruntime/test/hosting/integration_tests/README.MD b/onnxruntime/test/hosting/integration_tests/README.MD
deleted file mode 100644
index 2d308fe1425b2..0000000000000
--- a/onnxruntime/test/hosting/integration_tests/README.MD
+++ /dev/null
@@ -1,43 +0,0 @@
-# ONNX Runtime Hosting Application Integration Tests
-
-## Preparation
-
-Tests validation depends on protobuf generated *_pb2.py. So we need to have a sucessful hosting application build to have it generated in the build folder under `hosting_test` subfolder. The following instruction assume you are in the folder. Otherwise, tests will fail due to `ModuleNotFoundError`.
-
-## Functional Tests
-
-Functional test will be run when build with `--build_hosting --enable_hosting_tests`. To run it separately, here is the command line:
-
-```Bash
-/usr/bin/python3 ./test_main.py <hosting_app_path> <mnist_model_path> <test_data_path>
-```
-
-## Model Zoo Tests
-
-To run this set of tests, a prepared test data set need to be downloaded from [Azure Blob Storage](https://onnxhostingdev.blob.core.windows.net/testing/hosting_test_data_20190422.zip) and unzip to a folder, e.g. /home/foo/bar/model_zoo_test. It contains:
-
-* ONNX models from [ONNX Model Zoo](https://github.com/onnx/models) with opset 7/8/9.
-* HTTP request json and protobuf files
-* Expected response json and protobuf files
-
-If you only need the request and response data. Here is the [link](https://onnxhostingdev.blob.core.windows.net/testing/hosting_test_data_req_resp_only.zip) to download.
-
-To run the full model zoo tests, here is the command line:
-
-```Bash
-/usr/bin/python3 ./model_zoo_tests.py <hosting_app_path> <model_path> <test_data_path>
-```
-
-For example:
-
-```Bash
-/usr/bin/python3 ./model_zoo_tests.py /some/where/hosting_app /home/foo/bar/model_zoo_test /home/foo/bar/model_zoo_test
-```
-
-If those models are in different folder but in the same structure as the test data, you could also do 
-
-```Bash
-/usr/bin/python3 ./model_zoo_tests.py /some/where/hosting_app /home/my/models/ /home/foo/bar/model_zoo_test/
-```
-
-All tests are running in sequential order.
\ No newline at end of file
diff --git a/onnxruntime/test/server/integration_tests/README.MD b/onnxruntime/test/server/integration_tests/README.MD
new file mode 100644
index 0000000000000..141d531d1a9bc
--- /dev/null
+++ b/onnxruntime/test/server/integration_tests/README.MD
@@ -0,0 +1,43 @@
+# ONNX Runtime Server Integration Tests
+
+## Preparation
+
+Tests validation depends on protobuf generated *_pb2.py. So we need to have a sucessful server application build to have it generated in the build folder under `server_test` subfolder. The following instruction assume you are in the folder. Otherwise, tests will fail due to `ModuleNotFoundError`.
+
+## Functional Tests
+
+Functional test will be run when build with `--build_server --enable_server_tests`. To run it separately, here is the command line:
+
+```Bash
+/usr/bin/python3 ./test_main.py <server_app_path> <mnist_model_path> <test_data_path>
+```
+
+## Model Zoo Tests
+
+To run this set of tests, a prepared test data set need to be downloaded from [Azure Blob Storage](https://onnxserverdev.blob.core.windows.net/testing/server_test_data_20190422.zip) and unzip to a folder, e.g. /home/foo/bar/model_zoo_test. It contains:
+
+* ONNX models from [ONNX Model Zoo](https://github.com/onnx/models) with opset 7/8/9.
+* HTTP request json and protobuf files
+* Expected response json and protobuf files
+
+If you only need the request and response data. Here is the [link](https://onnxserverdev.blob.core.windows.net/testing/server_test_data_req_resp_only.zip) to download.
+
+To run the full model zoo tests, here is the command line:
+
+```Bash
+/usr/bin/python3 ./model_zoo_tests.py <server_app_path> <model_path> <test_data_path>
+```
+
+For example:
+
+```Bash
+/usr/bin/python3 ./model_zoo_tests.py /some/where/server_app /home/foo/bar/model_zoo_test /home/foo/bar/model_zoo_test
+```
+
+If those models are in different folder but in the same structure as the test data, you could also do 
+
+```Bash
+/usr/bin/python3 ./model_zoo_tests.py /some/where/server_app /home/my/models/ /home/foo/bar/model_zoo_test/
+```
+
+All tests are running in sequential order.
\ No newline at end of file
diff --git a/onnxruntime/test/hosting/integration_tests/function_tests.py b/onnxruntime/test/server/integration_tests/function_tests.py
similarity index 88%
rename from onnxruntime/test/hosting/integration_tests/function_tests.py
rename to onnxruntime/test/server/integration_tests/function_tests.py
index 76433cbfcb6f3..aabeb8bfc1ceb 100644
--- a/onnxruntime/test/hosting/integration_tests/function_tests.py
+++ b/onnxruntime/test/server/integration_tests/function_tests.py
@@ -17,28 +17,28 @@ class HttpJsonPayloadTests(unittest.TestCase):
     server_ip = '127.0.0.1'
     server_port = 54321
     url_pattern = 'http://{0}:{1}/v1/models/{2}/versions/{3}:predict'
-    hosting_app_path = ''
+    server_app_path = ''
     test_data_path = ''
     model_path = ''
     log_level = 'verbose'
-    hosting_app_proc = None
+    server_app_proc = None
     wait_server_ready_in_seconds = 1
 
     @classmethod
     def setUpClass(cls):
-        cmd = [cls.hosting_app_path, '--http_port', str(cls.server_port), '--model_path', os.path.join(cls.model_path, 'mnist.onnx'), '--log_level', cls.log_level]
-        print('Launching hosting app: [{0}]'.format(' '.join(cmd)))
-        cls.hosting_app_proc = subprocess.Popen(cmd)
-        print('Hosting app PID: {0}'.format(cls.hosting_app_proc.pid))
+        cmd = [cls.server_app_path, '--http_port', str(cls.server_port), '--model_path', os.path.join(cls.model_path, 'mnist.onnx'), '--log_level', cls.log_level]
+        print('Launching server app: [{0}]'.format(' '.join(cmd)))
+        cls.server_app_proc = subprocess.Popen(cmd)
+        print('Server app PID: {0}'.format(cls.server_app_proc.pid))
         print('Sleep {0} second(s) to wait for server initialization'.format(cls.wait_server_ready_in_seconds))
         time.sleep(cls.wait_server_ready_in_seconds)
 
 
     @classmethod
     def tearDownClass(cls):
-        print('Shutdown hosting app')
-        cls.hosting_app_proc.kill()
-        print('PID {0} has been killed: {1}'.format(cls.hosting_app_proc.pid, test_util.is_process_killed(cls.hosting_app_proc.pid)))
+        print('Shutdown server app')
+        cls.server_app_proc.kill()
+        print('PID {0} has been killed: {1}'.format(cls.server_app_proc.pid, test_util.is_process_killed(cls.server_app_proc.pid)))
 
 
     def test_mnist_happy_path(self):
@@ -193,28 +193,28 @@ class HttpProtobufPayloadTests(unittest.TestCase):
     server_ip = '127.0.0.1'
     server_port = 54321
     url_pattern = 'http://{0}:{1}/v1/models/{2}/versions/{3}:predict'
-    hosting_app_path = ''
+    server_app_path = ''
     test_data_path = ''
     model_path = ''
     log_level = 'verbose'
-    hosting_app_proc = None
+    server_app_proc = None
     wait_server_ready_in_seconds = 1
 
     @classmethod
     def setUpClass(cls):
-        cmd = [cls.hosting_app_path, '--http_port', str(cls.server_port), '--model_path', os.path.join(cls.model_path, 'mnist.onnx'), '--log_level', cls.log_level]
-        print('Launching hosting app: [{0}]'.format(' '.join(cmd)))
-        cls.hosting_app_proc = subprocess.Popen(cmd)
-        print('Hosting app PID: {0}'.format(cls.hosting_app_proc.pid))
+        cmd = [cls.server_app_path, '--http_port', str(cls.server_port), '--model_path', os.path.join(cls.model_path, 'mnist.onnx'), '--log_level', cls.log_level]
+        print('Launching server app: [{0}]'.format(' '.join(cmd)))
+        cls.server_app_proc = subprocess.Popen(cmd)
+        print('Server app PID: {0}'.format(cls.server_app_proc.pid))
         print('Sleep {0} second(s) to wait for server initialization'.format(cls.wait_server_ready_in_seconds))
         time.sleep(cls.wait_server_ready_in_seconds)
 
 
     @classmethod
     def tearDownClass(cls):
-        print('Shutdown hosting app')
-        cls.hosting_app_proc.kill()
-        print('PID {0} has been killed: {1}'.format(cls.hosting_app_proc.pid, test_util.is_process_killed(cls.hosting_app_proc.pid)))
+        print('Shutdown server app')
+        cls.server_app_proc.kill()
+        print('PID {0} has been killed: {1}'.format(cls.server_app_proc.pid, test_util.is_process_killed(cls.server_app_proc.pid)))
 
 
     def test_mnist_happy_path(self):
@@ -321,28 +321,28 @@ def test_any_accept_header(self):
 class HttpEndpointTests(unittest.TestCase):
     server_ip = '127.0.0.1'
     server_port = 54321
-    hosting_app_path = ''
+    server_app_path = ''
     test_data_path = ''
     model_path = ''
     log_level = 'verbose'
-    hosting_app_proc = None
+    server_app_proc = None
     wait_server_ready_in_seconds = 1
 
     @classmethod
     def setUpClass(cls):
-        cmd = [cls.hosting_app_path, '--http_port', str(cls.server_port), '--model_path', os.path.join(cls.model_path, 'mnist.onnx'), '--log_level', cls.log_level]
-        print('Launching hosting app: [{0}]'.format(' '.join(cmd)))
-        cls.hosting_app_proc = subprocess.Popen(cmd)
-        print('Hosting app PID: {0}'.format(cls.hosting_app_proc.pid))
+        cmd = [cls.server_app_path, '--http_port', str(cls.server_port), '--model_path', os.path.join(cls.model_path, 'mnist.onnx'), '--log_level', cls.log_level]
+        print('Launching server app: [{0}]'.format(' '.join(cmd)))
+        cls.server_app_proc = subprocess.Popen(cmd)
+        print('Server app PID: {0}'.format(cls.server_app_proc.pid))
         print('Sleep {0} second(s) to wait for server initialization'.format(cls.wait_server_ready_in_seconds))
         time.sleep(cls.wait_server_ready_in_seconds)
 
 
     @classmethod
     def tearDownClass(cls):
-        print('Shutdown hosting app')
-        cls.hosting_app_proc.kill()
-        print('PID {0} has been killed: {1}'.format(cls.hosting_app_proc.pid, test_util.is_process_killed(cls.hosting_app_proc.pid)))
+        print('Shutdown server app')
+        cls.server_app_proc.kill()
+        print('PID {0} has been killed: {1}'.format(cls.server_app_proc.pid, test_util.is_process_killed(cls.server_app_proc.pid)))
 
 
     def test_health_endpoint(self):
diff --git a/onnxruntime/test/hosting/integration_tests/model_zoo_data_prep.py b/onnxruntime/test/server/integration_tests/model_zoo_data_prep.py
similarity index 100%
rename from onnxruntime/test/hosting/integration_tests/model_zoo_data_prep.py
rename to onnxruntime/test/server/integration_tests/model_zoo_data_prep.py
diff --git a/onnxruntime/test/hosting/integration_tests/model_zoo_tests.py b/onnxruntime/test/server/integration_tests/model_zoo_tests.py
similarity index 88%
rename from onnxruntime/test/hosting/integration_tests/model_zoo_tests.py
rename to onnxruntime/test/server/integration_tests/model_zoo_tests.py
index 95f2e0e9a7438..c47a9c50b030d 100644
--- a/onnxruntime/test/hosting/integration_tests/model_zoo_tests.py
+++ b/onnxruntime/test/server/integration_tests/model_zoo_tests.py
@@ -11,7 +11,7 @@ class ModelZooTests(unittest.TestCase):
     server_ip = '127.0.0.1'
     server_port = 54321
     url_pattern = 'http://{0}:{1}/v1/models/{2}/versions/{3}:predict'
-    hosting_app_path = ''  # Required
+    server_app_path = ''  # Required
     log_level = 'verbose'
     server_ready_in_seconds = 10
     server_off_in_seconds = 100
@@ -55,11 +55,11 @@ def test_models_from_model_zoo(self):
 
         self.server_port = random.randint(30000, 40000)
         for model_path, data_paths in model_data_map.items():
-            hosting_app_proc = None
+            server_app_proc = None
             try:
-                cmd = [self.hosting_app_path, '--http_port', str(self.server_port), '--model_path', os.path.join(model_path, 'model.onnx'), '--log_level', self.log_level]
+                cmd = [self.server_app_path, '--http_port', str(self.server_port), '--model_path', os.path.join(model_path, 'model.onnx'), '--log_level', self.log_level]
                 test_util.test_log(cmd)
-                hosting_app_proc = test_util.launch_hosting_app(cmd, self.server_ip, self.server_port, self.server_ready_in_seconds)
+                server_app_proc = test_util.launch_server_app(cmd, self.server_ip, self.server_port, self.server_ready_in_seconds)
                
                 test_util.test_log('[{0}] Run tests...'.format(model_path))
                 for test in data_paths:
@@ -79,7 +79,7 @@ def test_models_from_model_zoo(self):
                     resp = test_util.make_http_request(url, pb_request_headers, request_payload)
                     test_util.pb_response_validation(self, resp, os.path.join(test, 'response.pb'))
             finally:
-                test_util.shutdown_hosting_app(hosting_app_proc, self.server_off_in_seconds)
+                test_util.shutdown_server_app(server_app_proc, self.server_off_in_seconds)
 
 
 if __name__ == '__main__':
@@ -89,7 +89,7 @@ def test_models_from_model_zoo(self):
 
     test_suites = []
     for tests in test_classes:
-        tests.hosting_app_path = sys.argv[1]
+        tests.server_app_path = sys.argv[1]
         tests.model_zoo_model_path = sys.argv[2]
         tests.model_zoo_test_data_path = sys.argv[3]
 
diff --git a/onnxruntime/test/hosting/integration_tests/test_main.py b/onnxruntime/test/server/integration_tests/test_main.py
similarity index 94%
rename from onnxruntime/test/hosting/integration_tests/test_main.py
rename to onnxruntime/test/server/integration_tests/test_main.py
index 5e34768a3f8ef..83ff585fb5f78 100644
--- a/onnxruntime/test/hosting/integration_tests/test_main.py
+++ b/onnxruntime/test/server/integration_tests/test_main.py
@@ -13,7 +13,7 @@
 
     test_suites = []
     for tests in test_classes:
-        tests.hosting_app_path = sys.argv[1]
+        tests.server_app_path = sys.argv[1]
         tests.model_path = sys.argv[2]
         tests.test_data_path = sys.argv[3]
         tests.server_port = random.randint(30000, 50000)
diff --git a/onnxruntime/test/hosting/integration_tests/test_util.py b/onnxruntime/test/server/integration_tests/test_util.py
similarity index 87%
rename from onnxruntime/test/hosting/integration_tests/test_util.py
rename to onnxruntime/test/server/integration_tests/test_util.py
index de1a926f41f3a..60840a873012b 100644
--- a/onnxruntime/test/hosting/integration_tests/test_util.py
+++ b/onnxruntime/test/server/integration_tests/test_util.py
@@ -70,24 +70,24 @@ def wait_service_up(server, port, timeout=1):
             return True
 
 
-def launch_hosting_app(cmd, server_ip, server_port, wait_server_ready_in_seconds):
-    test_log('Launching hosting app: [{0}]'.format(' '.join(cmd)))
-    hosting_app_proc = subprocess.Popen(cmd)
-    test_log('Hosting app PID: {0}'.format(hosting_app_proc.pid))
+def launch_server_app(cmd, server_ip, server_port, wait_server_ready_in_seconds):
+    test_log('Launching server app: [{0}]'.format(' '.join(cmd)))
+    server_app_proc = subprocess.Popen(cmd)
+    test_log('Server app PID: {0}'.format(server_app_proc.pid))
     test_log('Wait up to {0} second(s) for server initialization'.format(wait_server_ready_in_seconds))
     wait_service_up(server_ip, server_port, wait_server_ready_in_seconds)
 
-    return hosting_app_proc
+    return server_app_proc
 
 
-def shutdown_hosting_app(hosting_app_proc, wait_for_server_off_in_seconds):
-    if hosting_app_proc is not None:
-        test_log('Shutdown hosting app')
-        hosting_app_proc.kill()
+def shutdown_server_app(server_app_proc, wait_for_server_off_in_seconds):
+    if server_app_proc is not None:
+        test_log('Shutdown server app')
+        server_app_proc.kill()
 
-        while not is_process_killed(hosting_app_proc.pid):
-            hosting_app_proc.wait(timeout=wait_for_server_off_in_seconds)
-            test_log('PID {0} has been killed: {1}'.format(hosting_app_proc.pid, is_process_killed(hosting_app_proc.pid)))
+        while not is_process_killed(server_app_proc.pid):
+            server_app_proc.wait(timeout=wait_for_server_off_in_seconds)
+            test_log('PID {0} has been killed: {1}'.format(server_app_proc.pid, is_process_killed(server_app_proc.pid)))
 
         # Additional sleep to make sure the resource has been freed.
         time.sleep(1)
diff --git a/onnxruntime/test/hosting/unit_tests/converter_tests.cc b/onnxruntime/test/server/unit_tests/converter_tests.cc
similarity index 88%
rename from onnxruntime/test/hosting/unit_tests/converter_tests.cc
rename to onnxruntime/test/server/unit_tests/converter_tests.cc
index a470cc12e6258..e5f2d9642f2f6 100644
--- a/onnxruntime/test/hosting/unit_tests/converter_tests.cc
+++ b/onnxruntime/test/server/unit_tests/converter_tests.cc
@@ -8,10 +8,10 @@
 #include "core/framework/allocatormgr.h"
 #include "test/framework/test_utils.h"
 #include "test/test_environment.h"
-#include "hosting/converter.h"
+#include "server/converter.h"
 
 namespace onnxruntime {
-namespace hosting {
+namespace server {
 namespace test {
 
 void CreateMLValueBool(AllocatorPtr alloc, const std::vector<int64_t>& dims, const bool* value, MLValue* p_mlvalue);
@@ -26,63 +26,63 @@ TEST(MLDataTypeToTensorProtoDataTypeTests, MLDataTypeToTensorProtoDataTypeTests)
   auto logger = std::make_unique<onnxruntime::logging::Logger>(::onnxruntime::test::DefaultLoggingManager().DefaultLogger());
 
   MLDataType ml_data_type = DataTypeImpl::GetType<float>();
-  onnx::TensorProto_DataType result = onnxruntime::hosting::MLDataTypeToTensorProtoDataType(ml_data_type);
+  onnx::TensorProto_DataType result = onnxruntime::server::MLDataTypeToTensorProtoDataType(ml_data_type);
   EXPECT_EQ(result, onnx::TensorProto_DataType_FLOAT);
 
   ml_data_type = DataTypeImpl::GetType<onnxruntime::MLFloat16>();
-  result = onnxruntime::hosting::MLDataTypeToTensorProtoDataType(ml_data_type);
+  result = onnxruntime::server::MLDataTypeToTensorProtoDataType(ml_data_type);
   EXPECT_EQ(result, onnx::TensorProto_DataType_FLOAT16);
 
   ml_data_type = DataTypeImpl::GetType<onnxruntime::BFloat16>();
-  result = onnxruntime::hosting::MLDataTypeToTensorProtoDataType(ml_data_type);
+  result = onnxruntime::server::MLDataTypeToTensorProtoDataType(ml_data_type);
   EXPECT_EQ(result, onnx::TensorProto_DataType_BFLOAT16);
 
   ml_data_type = DataTypeImpl::GetType<double>();
-  result = onnxruntime::hosting::MLDataTypeToTensorProtoDataType(ml_data_type);
+  result = onnxruntime::server::MLDataTypeToTensorProtoDataType(ml_data_type);
   EXPECT_EQ(result, onnx::TensorProto_DataType_DOUBLE);
 
   ml_data_type = DataTypeImpl::GetType<uint8_t>();
-  result = onnxruntime::hosting::MLDataTypeToTensorProtoDataType(ml_data_type);
+  result = onnxruntime::server::MLDataTypeToTensorProtoDataType(ml_data_type);
   EXPECT_EQ(result, onnx::TensorProto_DataType_UINT8);
 
   ml_data_type = DataTypeImpl::GetType<int8_t>();
-  result = onnxruntime::hosting::MLDataTypeToTensorProtoDataType(ml_data_type);
+  result = onnxruntime::server::MLDataTypeToTensorProtoDataType(ml_data_type);
   EXPECT_EQ(result, onnx::TensorProto_DataType_INT8);
 
   ml_data_type = DataTypeImpl::GetType<uint16_t>();
-  result = onnxruntime::hosting::MLDataTypeToTensorProtoDataType(ml_data_type);
+  result = onnxruntime::server::MLDataTypeToTensorProtoDataType(ml_data_type);
   EXPECT_EQ(result, onnx::TensorProto_DataType_UINT16);
 
   ml_data_type = DataTypeImpl::GetType<int16_t>();
-  result = onnxruntime::hosting::MLDataTypeToTensorProtoDataType(ml_data_type);
+  result = onnxruntime::server::MLDataTypeToTensorProtoDataType(ml_data_type);
   EXPECT_EQ(result, onnx::TensorProto_DataType_INT16);
 
   ml_data_type = DataTypeImpl::GetType<uint32_t>();
-  result = onnxruntime::hosting::MLDataTypeToTensorProtoDataType(ml_data_type);
+  result = onnxruntime::server::MLDataTypeToTensorProtoDataType(ml_data_type);
   EXPECT_EQ(result, onnx::TensorProto_DataType_UINT32);
 
   ml_data_type = DataTypeImpl::GetType<int32_t>();
-  result = onnxruntime::hosting::MLDataTypeToTensorProtoDataType(ml_data_type);
+  result = onnxruntime::server::MLDataTypeToTensorProtoDataType(ml_data_type);
   EXPECT_EQ(result, onnx::TensorProto_DataType_INT32);
 
   ml_data_type = DataTypeImpl::GetType<uint64_t>();
-  result = onnxruntime::hosting::MLDataTypeToTensorProtoDataType(ml_data_type);
+  result = onnxruntime::server::MLDataTypeToTensorProtoDataType(ml_data_type);
   EXPECT_EQ(result, onnx::TensorProto_DataType_UINT64);
 
   ml_data_type = DataTypeImpl::GetType<int64_t>();
-  result = onnxruntime::hosting::MLDataTypeToTensorProtoDataType(ml_data_type);
+  result = onnxruntime::server::MLDataTypeToTensorProtoDataType(ml_data_type);
   EXPECT_EQ(result, onnx::TensorProto_DataType_INT64);
 
   ml_data_type = DataTypeImpl::GetType<std::string>();
-  result = onnxruntime::hosting::MLDataTypeToTensorProtoDataType(ml_data_type);
+  result = onnxruntime::server::MLDataTypeToTensorProtoDataType(ml_data_type);
   EXPECT_EQ(result, onnx::TensorProto_DataType_STRING);
 
   ml_data_type = DataTypeImpl::GetType<bool>();
-  result = onnxruntime::hosting::MLDataTypeToTensorProtoDataType(ml_data_type);
+  result = onnxruntime::server::MLDataTypeToTensorProtoDataType(ml_data_type);
   EXPECT_EQ(result, onnx::TensorProto_DataType_BOOL);
 
   ml_data_type = DataTypeImpl::GetTensorType<bool>();
-  result = onnxruntime::hosting::MLDataTypeToTensorProtoDataType(ml_data_type);
+  result = onnxruntime::server::MLDataTypeToTensorProtoDataType(ml_data_type);
   EXPECT_EQ(result, onnx::TensorProto_DataType_UNDEFINED);
 }
 
@@ -95,7 +95,7 @@ TEST(MLValueToTensorProtoTests, FloatToRaw) {
   onnxruntime::test::CreateMLValue<float>(TestCPUExecutionProvider()->GetAllocator(0, OrtMemTypeDefault), dims_mul_x, values_mul_x, &ml_value);
 
   onnx::TensorProto tp;
-  common::Status status = onnxruntime::hosting::MLValueToTensorProto(ml_value, /* using_raw_data */ true, std::move(logger), tp);
+  common::Status status = onnxruntime::server::MLValueToTensorProto(ml_value, /* using_raw_data */ true, std::move(logger), tp);
   EXPECT_TRUE(status.IsOK());
 
   // Verify data type
@@ -134,7 +134,7 @@ TEST(MLValueToTensorProtoTests, FloatToFloatData) {
   onnxruntime::test::CreateMLValue<float>(TestCPUExecutionProvider()->GetAllocator(0, OrtMemTypeDefault), dims_mul_x, values_mul_x, &ml_value);
 
   onnx::TensorProto tp;
-  common::Status status = onnxruntime::hosting::MLValueToTensorProto(ml_value, /* using_raw_data */ false, std::move(logger), tp);
+  common::Status status = onnxruntime::server::MLValueToTensorProto(ml_value, /* using_raw_data */ false, std::move(logger), tp);
   EXPECT_TRUE(status.IsOK());
 
   // Verify data type
@@ -169,7 +169,7 @@ TEST(MLValueToTensorProtoTests, Int32ToRaw) {
   onnxruntime::test::CreateMLValue<int32_t>(TestCPUExecutionProvider()->GetAllocator(0, OrtMemTypeDefault), dims_mul_x, values_mul_x, &ml_value);
 
   onnx::TensorProto tp;
-  common::Status status = onnxruntime::hosting::MLValueToTensorProto(ml_value, /* using_raw_data */ true, std::move(logger), tp);
+  common::Status status = onnxruntime::server::MLValueToTensorProto(ml_value, /* using_raw_data */ true, std::move(logger), tp);
   EXPECT_TRUE(status.IsOK());
 
   // Verify data type
@@ -208,7 +208,7 @@ TEST(MLValueToTensorProtoTests, Int32ToInt32Data) {
   onnxruntime::test::CreateMLValue<int32_t>(TestCPUExecutionProvider()->GetAllocator(0, OrtMemTypeDefault), dims_mul_x, values_mul_x, &ml_value);
 
   onnx::TensorProto tp;
-  common::Status status = onnxruntime::hosting::MLValueToTensorProto(ml_value, /* using_raw_data */ false, std::move(logger), tp);
+  common::Status status = onnxruntime::server::MLValueToTensorProto(ml_value, /* using_raw_data */ false, std::move(logger), tp);
   EXPECT_TRUE(status.IsOK());
 
   // Verify data type
@@ -243,7 +243,7 @@ TEST(MLValueToTensorProtoTests, UInt8ToRaw) {
   onnxruntime::test::CreateMLValue<uint8_t>(TestCPUExecutionProvider()->GetAllocator(0, OrtMemTypeDefault), dims_mul_x, values_mul_x, &ml_value);
 
   onnx::TensorProto tp;
-  common::Status status = onnxruntime::hosting::MLValueToTensorProto(ml_value, /* using_raw_data */ true, std::move(logger), tp);
+  common::Status status = onnxruntime::server::MLValueToTensorProto(ml_value, /* using_raw_data */ true, std::move(logger), tp);
   EXPECT_TRUE(status.IsOK());
 
   // Verify data type
@@ -282,7 +282,7 @@ TEST(MLValueToTensorProtoTests, UInt8ToInt32Data) {
   onnxruntime::test::CreateMLValue<uint8_t>(TestCPUExecutionProvider()->GetAllocator(0, OrtMemTypeDefault), dims_mul_x, values_mul_x, &ml_value);
 
   onnx::TensorProto tp;
-  common::Status status = onnxruntime::hosting::MLValueToTensorProto(ml_value, /* using_raw_data */ false, std::move(logger), tp);
+  common::Status status = onnxruntime::server::MLValueToTensorProto(ml_value, /* using_raw_data */ false, std::move(logger), tp);
   EXPECT_TRUE(status.IsOK());
 
   // Verify data type
@@ -319,7 +319,7 @@ TEST(MLValueToTensorProtoTests, Int8ToRaw) {
   onnxruntime::test::CreateMLValue<int8_t>(TestCPUExecutionProvider()->GetAllocator(0, OrtMemTypeDefault), dims_mul_x, values_mul_x, &ml_value);
 
   onnx::TensorProto tp;
-  common::Status status = onnxruntime::hosting::MLValueToTensorProto(ml_value, /* using_raw_data */ true, std::move(logger), tp);
+  common::Status status = onnxruntime::server::MLValueToTensorProto(ml_value, /* using_raw_data */ true, std::move(logger), tp);
   EXPECT_TRUE(status.IsOK());
 
   // Verify data type
@@ -358,7 +358,7 @@ TEST(MLValueToTensorProtoTests, Int8ToInt32Data) {
   onnxruntime::test::CreateMLValue<int8_t>(TestCPUExecutionProvider()->GetAllocator(0, OrtMemTypeDefault), dims_mul_x, values_mul_x, &ml_value);
 
   onnx::TensorProto tp;
-  common::Status status = onnxruntime::hosting::MLValueToTensorProto(ml_value, /* using_raw_data */ false, std::move(logger), tp);
+  common::Status status = onnxruntime::server::MLValueToTensorProto(ml_value, /* using_raw_data */ false, std::move(logger), tp);
   EXPECT_TRUE(status.IsOK());
 
   // Verify data type
@@ -395,7 +395,7 @@ TEST(MLValueToTensorProtoTests, UInt16ToRaw) {
   onnxruntime::test::CreateMLValue<uint16_t>(TestCPUExecutionProvider()->GetAllocator(0, OrtMemTypeDefault), dims_mul_x, values_mul_x, &ml_value);
 
   onnx::TensorProto tp;
-  common::Status status = onnxruntime::hosting::MLValueToTensorProto(ml_value, /* using_raw_data */ true, std::move(logger), tp);
+  common::Status status = onnxruntime::server::MLValueToTensorProto(ml_value, /* using_raw_data */ true, std::move(logger), tp);
   EXPECT_TRUE(status.IsOK());
 
   // Verify data type
@@ -434,7 +434,7 @@ TEST(MLValueToTensorProtoTests, UInt16ToInt32Data) {
   onnxruntime::test::CreateMLValue<uint16_t>(TestCPUExecutionProvider()->GetAllocator(0, OrtMemTypeDefault), dims_mul_x, values_mul_x, &ml_value);
 
   onnx::TensorProto tp;
-  common::Status status = onnxruntime::hosting::MLValueToTensorProto(ml_value, /* using_raw_data */ false, std::move(logger), tp);
+  common::Status status = onnxruntime::server::MLValueToTensorProto(ml_value, /* using_raw_data */ false, std::move(logger), tp);
   EXPECT_TRUE(status.IsOK());
 
   // Verify data type
@@ -471,7 +471,7 @@ TEST(MLValueToTensorProtoTests, Int16ToRaw) {
   onnxruntime::test::CreateMLValue<int16_t>(TestCPUExecutionProvider()->GetAllocator(0, OrtMemTypeDefault), dims_mul_x, values_mul_x, &ml_value);
 
   onnx::TensorProto tp;
-  common::Status status = onnxruntime::hosting::MLValueToTensorProto(ml_value, /* using_raw_data */ true, std::move(logger), tp);
+  common::Status status = onnxruntime::server::MLValueToTensorProto(ml_value, /* using_raw_data */ true, std::move(logger), tp);
   EXPECT_TRUE(status.IsOK());
 
   // Verify data type
@@ -510,7 +510,7 @@ TEST(MLValueToTensorProtoTests, Int16ToInt32Data) {
   onnxruntime::test::CreateMLValue<int16_t>(TestCPUExecutionProvider()->GetAllocator(0, OrtMemTypeDefault), dims_mul_x, values_mul_x, &ml_value);
 
   onnx::TensorProto tp;
-  common::Status status = onnxruntime::hosting::MLValueToTensorProto(ml_value, /* using_raw_data */ false, std::move(logger), tp);
+  common::Status status = onnxruntime::server::MLValueToTensorProto(ml_value, /* using_raw_data */ false, std::move(logger), tp);
   EXPECT_TRUE(status.IsOK());
 
   // Verify data type
@@ -547,7 +547,7 @@ TEST(MLValueToTensorProtoTests, BoolToRaw) {
   CreateMLValueBool(TestCPUExecutionProvider()->GetAllocator(0, OrtMemTypeDefault), dims_mul_x, values_mul_x, &ml_value);
 
   onnx::TensorProto tp;
-  common::Status status = onnxruntime::hosting::MLValueToTensorProto(ml_value, /* using_raw_data */ true, std::move(logger), tp);
+  common::Status status = onnxruntime::server::MLValueToTensorProto(ml_value, /* using_raw_data */ true, std::move(logger), tp);
   EXPECT_TRUE(status.IsOK());
 
   // Verify data type
@@ -586,7 +586,7 @@ TEST(MLValueToTensorProtoTests, BoolToInt32Data) {
   CreateMLValueBool(TestCPUExecutionProvider()->GetAllocator(0, OrtMemTypeDefault), dims_mul_x, values_mul_x, &ml_value);
 
   onnx::TensorProto tp;
-  common::Status status = onnxruntime::hosting::MLValueToTensorProto(ml_value, /* using_raw_data */ false, std::move(logger), tp);
+  common::Status status = onnxruntime::server::MLValueToTensorProto(ml_value, /* using_raw_data */ false, std::move(logger), tp);
   EXPECT_TRUE(status.IsOK());
 
   // Verify data type
@@ -629,7 +629,7 @@ TEST(MLValueToTensorProtoTests, Float16ToRaw) {
   onnxruntime::test::CreateMLValue<onnxruntime::MLFloat16>(TestCPUExecutionProvider()->GetAllocator(0, OrtMemTypeDefault), dims_mul_x, values_mul_x, &ml_value);
 
   onnx::TensorProto tp;
-  common::Status status = onnxruntime::hosting::MLValueToTensorProto(ml_value, /* using_raw_data */ true, std::move(logger), tp);
+  common::Status status = onnxruntime::server::MLValueToTensorProto(ml_value, /* using_raw_data */ true, std::move(logger), tp);
   EXPECT_TRUE(status.IsOK());
 
   // Verify data type
@@ -674,7 +674,7 @@ TEST(MLValueToTensorProtoTests, FloatToInt32Data) {
   onnxruntime::test::CreateMLValue<onnxruntime::MLFloat16>(TestCPUExecutionProvider()->GetAllocator(0, OrtMemTypeDefault), dims_mul_x, values_mul_x, &ml_value);
 
   onnx::TensorProto tp;
-  common::Status status = onnxruntime::hosting::MLValueToTensorProto(ml_value, /* using_raw_data */ false, std::move(logger), tp);
+  common::Status status = onnxruntime::server::MLValueToTensorProto(ml_value, /* using_raw_data */ false, std::move(logger), tp);
   EXPECT_TRUE(status.IsOK());
 
   // Verify data type
@@ -717,7 +717,7 @@ TEST(MLValueToTensorProtoTests, BFloat16ToRaw) {
   onnxruntime::test::CreateMLValue<onnxruntime::BFloat16>(TestCPUExecutionProvider()->GetAllocator(0, OrtMemTypeDefault), dims_mul_x, values_mul_x, &ml_value);
 
   onnx::TensorProto tp;
-  common::Status status = onnxruntime::hosting::MLValueToTensorProto(ml_value, /* using_raw_data */ true, std::move(logger), tp);
+  common::Status status = onnxruntime::server::MLValueToTensorProto(ml_value, /* using_raw_data */ true, std::move(logger), tp);
   EXPECT_TRUE(status.IsOK());
 
   // Verify data type
@@ -762,7 +762,7 @@ TEST(MLValueToTensorProtoTests, BFloatToInt32Data) {
   onnxruntime::test::CreateMLValue<onnxruntime::BFloat16>(TestCPUExecutionProvider()->GetAllocator(0, OrtMemTypeDefault), dims_mul_x, values_mul_x, &ml_value);
 
   onnx::TensorProto tp;
-  common::Status status = onnxruntime::hosting::MLValueToTensorProto(ml_value, /* using_raw_data */ false, std::move(logger), tp);
+  common::Status status = onnxruntime::server::MLValueToTensorProto(ml_value, /* using_raw_data */ false, std::move(logger), tp);
   EXPECT_TRUE(status.IsOK());
 
   // Verify data type
@@ -805,7 +805,7 @@ TEST(MLValueToTensorProtoTests, StringToStringData) {
   }
 
   onnx::TensorProto tp;
-  common::Status status = onnxruntime::hosting::MLValueToTensorProto(ml_value, /* using_raw_data */ false, std::move(logger), tp);
+  common::Status status = onnxruntime::server::MLValueToTensorProto(ml_value, /* using_raw_data */ false, std::move(logger), tp);
   EXPECT_TRUE(status.IsOK());
 
   // Verify data type
@@ -841,7 +841,7 @@ TEST(MLValueToTensorProtoTests, Int64ToRaw) {
   onnxruntime::test::CreateMLValue<int64_t>(TestCPUExecutionProvider()->GetAllocator(0, OrtMemTypeDefault), dims_mul_x, values_mul_x, &ml_value);
 
   onnx::TensorProto tp;
-  common::Status status = onnxruntime::hosting::MLValueToTensorProto(ml_value, /* using_raw_data */ true, std::move(logger), tp);
+  common::Status status = onnxruntime::server::MLValueToTensorProto(ml_value, /* using_raw_data */ true, std::move(logger), tp);
   EXPECT_TRUE(status.IsOK());
 
   // Verify data type
@@ -880,7 +880,7 @@ TEST(MLValueToTensorProtoTests, Int64ToInt64Data) {
   onnxruntime::test::CreateMLValue<int64_t>(TestCPUExecutionProvider()->GetAllocator(0, OrtMemTypeDefault), dims_mul_x, values_mul_x, &ml_value);
 
   onnx::TensorProto tp;
-  common::Status status = onnxruntime::hosting::MLValueToTensorProto(ml_value, /* using_raw_data */ false, std::move(logger), tp);
+  common::Status status = onnxruntime::server::MLValueToTensorProto(ml_value, /* using_raw_data */ false, std::move(logger), tp);
   EXPECT_TRUE(status.IsOK());
 
   // Verify data type
@@ -915,7 +915,7 @@ TEST(MLValueToTensorProtoTests, UInt32ToRaw) {
   onnxruntime::test::CreateMLValue<uint32_t>(TestCPUExecutionProvider()->GetAllocator(0, OrtMemTypeDefault), dims_mul_x, values_mul_x, &ml_value);
 
   onnx::TensorProto tp;
-  common::Status status = onnxruntime::hosting::MLValueToTensorProto(ml_value, /* using_raw_data */ true, std::move(logger), tp);
+  common::Status status = onnxruntime::server::MLValueToTensorProto(ml_value, /* using_raw_data */ true, std::move(logger), tp);
   EXPECT_TRUE(status.IsOK());
 
   // Verify data type
@@ -954,7 +954,7 @@ TEST(MLValueToTensorProtoTests, UInt32ToUint64Data) {
   onnxruntime::test::CreateMLValue<uint32_t>(TestCPUExecutionProvider()->GetAllocator(0, OrtMemTypeDefault), dims_mul_x, values_mul_x, &ml_value);
 
   onnx::TensorProto tp;
-  common::Status status = onnxruntime::hosting::MLValueToTensorProto(ml_value, /* using_raw_data */ false, std::move(logger), tp);
+  common::Status status = onnxruntime::server::MLValueToTensorProto(ml_value, /* using_raw_data */ false, std::move(logger), tp);
   EXPECT_TRUE(status.IsOK());
 
   // Verify data type
@@ -992,7 +992,7 @@ TEST(MLValueToTensorProtoTests, UInt64ToRaw) {
   onnxruntime::test::CreateMLValue<uint64_t>(TestCPUExecutionProvider()->GetAllocator(0, OrtMemTypeDefault), dims_mul_x, values_mul_x, &ml_value);
 
   onnx::TensorProto tp;
-  common::Status status = onnxruntime::hosting::MLValueToTensorProto(ml_value, /* using_raw_data */ true, std::move(logger), tp);
+  common::Status status = onnxruntime::server::MLValueToTensorProto(ml_value, /* using_raw_data */ true, std::move(logger), tp);
   EXPECT_TRUE(status.IsOK());
 
   // Verify data type
@@ -1031,7 +1031,7 @@ TEST(MLValueToTensorProtoTests, UInt64ToInt64Data) {
   onnxruntime::test::CreateMLValue<uint64_t>(TestCPUExecutionProvider()->GetAllocator(0, OrtMemTypeDefault), dims_mul_x, values_mul_x, &ml_value);
 
   onnx::TensorProto tp;
-  common::Status status = onnxruntime::hosting::MLValueToTensorProto(ml_value, /* using_raw_data */ false, std::move(logger), tp);
+  common::Status status = onnxruntime::server::MLValueToTensorProto(ml_value, /* using_raw_data */ false, std::move(logger), tp);
   EXPECT_TRUE(status.IsOK());
 
   // Verify data type
@@ -1066,7 +1066,7 @@ TEST(MLValueToTensorProtoTests, DoubleToRaw) {
   onnxruntime::test::CreateMLValue<double>(TestCPUExecutionProvider()->GetAllocator(0, OrtMemTypeDefault), dims_mul_x, values_mul_x, &ml_value);
 
   onnx::TensorProto tp;
-  common::Status status = onnxruntime::hosting::MLValueToTensorProto(ml_value, /* using_raw_data */ true, std::move(logger), tp);
+  common::Status status = onnxruntime::server::MLValueToTensorProto(ml_value, /* using_raw_data */ true, std::move(logger), tp);
   EXPECT_TRUE(status.IsOK());
 
   // Verify data type
@@ -1105,7 +1105,7 @@ TEST(MLValueToTensorProtoTests, DoubleToInt64Data) {
   onnxruntime::test::CreateMLValue<double>(TestCPUExecutionProvider()->GetAllocator(0, OrtMemTypeDefault), dims_mul_x, values_mul_x, &ml_value);
 
   onnx::TensorProto tp;
-  common::Status status = onnxruntime::hosting::MLValueToTensorProto(ml_value, /* using_raw_data */ false, std::move(logger), tp);
+  common::Status status = onnxruntime::server::MLValueToTensorProto(ml_value, /* using_raw_data */ false, std::move(logger), tp);
   EXPECT_TRUE(status.IsOK());
 
   // Verify data type
@@ -1147,5 +1147,5 @@ void CreateMLValueBool(AllocatorPtr alloc,
 }
 
 }  // namespace test
-}  // namespace hosting
+}  // namespace server
 }  // namespace onnxruntime
\ No newline at end of file
diff --git a/onnxruntime/test/hosting/unit_tests/http_routes_tests.cc b/onnxruntime/test/server/unit_tests/http_routes_tests.cc
similarity index 98%
rename from onnxruntime/test/hosting/unit_tests/http_routes_tests.cc
rename to onnxruntime/test/server/unit_tests/http_routes_tests.cc
index 9acc32adce9c9..dbb955050165c 100644
--- a/onnxruntime/test/hosting/unit_tests/http_routes_tests.cc
+++ b/onnxruntime/test/server/unit_tests/http_routes_tests.cc
@@ -4,10 +4,10 @@
 #include <iostream>
 
 #include "gtest/gtest.h"
-#include "hosting/http/core/routes.h"
+#include "server/http/core/routes.h"
 
 namespace onnxruntime {
-namespace hosting {
+namespace server {
 namespace test {
 
 using test_data = std::tuple<http::verb, std::string, std::string, std::string, std::string, http::status>;
@@ -105,5 +105,5 @@ void run_route(const std::string& pattern, http::verb method, const std::vector<
 }
 
 }  // namespace test
-}  // namespace hosting
+}  // namespace server
 }  // namespace onnxruntime
diff --git a/onnxruntime/test/hosting/unit_tests/json_handling_tests.cc b/onnxruntime/test/server/unit_tests/json_handling_tests.cc
similarity index 93%
rename from onnxruntime/test/hosting/unit_tests/json_handling_tests.cc
rename to onnxruntime/test/server/unit_tests/json_handling_tests.cc
index 3bb62a3945eb7..0bd2d16e45795 100644
--- a/onnxruntime/test/hosting/unit_tests/json_handling_tests.cc
+++ b/onnxruntime/test/server/unit_tests/json_handling_tests.cc
@@ -7,33 +7,33 @@
 #include "gtest/gtest.h"
 
 #include "predict.pb.h"
-#include "hosting/http/json_handling.h"
+#include "server/http/json_handling.h"
 
 namespace onnxruntime {
-namespace hosting {
+namespace server {
 namespace test {
 namespace protobufutil = google::protobuf::util;
 
 TEST(JsonDeserializationTests, HappyPath) {
   std::string input_json = R"({"inputs":{"Input3":{"dims":["1","1","28","28"],"dataType":1,"rawData":"AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACAPwAAQEAAAAAAAAAAAAAAgEAAAABAAAAAAAAAMEEAAAAAAAAAAAAAYEEAAIA/AAAAAAAAmEEAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAQEEAAAAAAAAAAAAA4EAAAAAAAACAPwAAIEEAAAAAAAAAQAAAAEAAAIBBAAAAAAAAQEAAAEBAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA4EAAAABBAAAAAAAAAEEAAAAAAAAAAAAAAEEAAAAAAAAAAAAAmEEAAAAAAAAAAAAAgD8AAKhBAAAAAAAAgEAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAgD8AAAAAAAAAAAAAgD8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAMEEAAAAAAAAAAAAAIEEAAEBAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABQQQAAAAAAAHBBAAAgQQAA0EEAAAhCAACIQQAAmkIAADVDAAAyQwAADEIAAIBAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAWQwAAfkMAAHpDAAB7QwAAc0MAAHxDAAB8QwAAf0MAADRCAADAQAAAAAAAAKBAAAAAAAAAEEEAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAOBAAACQQgAATUMAAH9DAABuQwAAc0MAAH9DAAB+QwAAe0MAAHhDAABJQwAARkMAAGRCAAAAAAAAmEEAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAWkMAAH9DAABxQwAAf0MAAHlDAAB6QwAAe0MAAHpDAAB/QwAAf0MAAHJDAABgQwAAREIAAAAAAABAQQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAgD8AAABAAABAQAAAAEAAAABAAACAPwAAAAAAAIJCAABkQwAAf0MAAH5DAAB0QwAA7kIAAAhCAAAkQgAA3EIAAHpDAAB/QwAAeEMAAPhCAACgQQAAAAAAAAAAAAAAAAAAAAAAAAAAAACAPwAAgD8AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEBBAAAAAAAAeEIAAM5CAADiQgAA6kIAAAhCAAAAAAAAAAAAAAAAAABIQwAAdEMAAH9DAAB/QwAAAAAAAEBBAAAAAAAAAAAAAAAAAAAAAAAAAEAAAIA/AAAAAAAAAAAAAAAAAAAAAAAAgD8AAABAAAAAAAAAAAAAAABAAACAQAAAAAAAADBBAAAAAAAA4EAAAMBAAAAAAAAAlkIAAHRDAAB/QwAAf0MAAIBAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAIA/AAAAQAAAQEAAAIBAAACAQAAAAAAAAGBBAAAAAAAAAAAAAAAAAAAQQQAAAAAAAABAAAAAAAAAAAAAAAhCAAB/QwAAf0MAAH1DAAAgQQAAIEEAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAIA/AAAAQAAAQEAAAABAAAAAAAAAAAAAAEBAAAAAQAAAAAAAAFBBAAAwQQAAAAAAAAAAAAAAAAAAwEAAAEBBAADGQgAAf0MAAH5DAAB4QwAAcEEAAEBBAAAAAAAAAAAAAAAAAAAAAAAAAAAAAIA/AACAPwAAgD8AAAAAAAAAAAAAAAAAAAAAAACAPwAAgD8AAAAAAAAAAAAAoEAAAMBAAAAwQQAAAAAAAAAAAACIQQAAOEMAAHdDAAB/QwAAc0MAAFBBAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEBAAAAAQAAAAAAAAAAAAAAAAAAAAAAAAABAAACAQAAAgEAAAAAAAAAwQQAAAAAAAExCAAC8QgAAqkIAAKBAAACgQAAAyEEAAHZDAAB2QwAAf0MAAFBDAAAAAAAAEEEAAAAAAAAAAAAAAAAAAAAAAACAQAAAgD8AAAAAAAAAAAAAgD8AAOBAAABwQQAAmEEAAMZCAADOQgAANkMAAD1DAABtQwAAfUMAAHxDAAA/QwAAPkMAAGNDAABzQwAAfEMAAFJDAACQQQAA4EAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAIBAAAAAAAAAAAAAAABCAADaQgAAOUMAAHdDAAB/QwAAckMAAH9DAAB0QwAAf0MAAH9DAAByQwAAe0MAAH9DAABwQwAAf0MAAH9DAABaQwAA+EIAABBBAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABAAAAAAAAAAAAAAAAAAAD+QgAAf0MAAGtDAAB/QwAAf0MAAHdDAABlQwAAVEMAAHJDAAB6QwAAf0MAAH9DAAB4QwAAf0MAAH1DAAB5QwAAf0MAAHNDAAAqQwAAQEEAAAAAAAAAAAAAAAAAAAAAAAAAAAAAMEEAAAAAAAAQQQAAfUMAAH9DAAB/QwAAaUMAAEpDAACqQgAAAAAAAFRCAABEQwAAbkMAAH9DAABjQwAAbkMAAA5DAADaQgAAQUMAAH9DAABwQwAAf0MAADRDAAAAAAAAAAAAAAAAAAAAAAAAwEAAAAAAAACwQQAAgD8AAHVDAABzQwAAfkMAAH9DAABZQwAAa0MAAGJDAABVQwAAdEMAAHtDAAB/QwAAb0MAAJpCAAAAAAAAAAAAAKBBAAA2QwAAd0MAAG9DAABzQwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAIBAAAAlQwAAe0MAAH9DAAB1QwAAf0MAAHJDAAB9QwAAekMAAH9DAABFQwAA1kIAAGxCAAAAAAAAkEEAAABAAADAQAAAAAAAAFhCAAB/QwAAHkMAAAAAAAAAAAAAAAAAAAAAAAAAAAAAwEEAAAAAAAAAAAAAwEAAAAhCAAAnQwAAQkMAADBDAAA3QwAAJEMAADBCAAAAQAAAIEEAAMBAAADAQAAAAAAAAAAAAACgQAAAAAAAAIA/AAAAAAAAYEEAAABAAAAAAAAAAAAAAAAAAAAAAAAAIEEAAAAAAABgQQAAAAAAAEBBAAAAAAAAoEAAAAAAAACAPwAAAAAAAMBAAAAAAAAA4EAAAAAAAAAAAAAAAAAAAABBAAAAAAAAIEEAAAAAAACgQAAAAAAAAAAAAAAgQQAAAAAAAAAAAAAAAAAAAAAAAAAAAABgQQAAAAAAAIBAAAAAAAAAAAAAAMhBAAAAAAAAAAAAABBBAAAAAAAAAAAAABBBAAAAAAAAMEEAAAAAAACAPwAAAAAAAAAAAAAAQAAAAAAAAAAAAADgQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA=="}},"outputFilter":["Plus214_Output_0"]})";
-  onnxruntime::hosting::PredictRequest request;
-  protobufutil::Status status = onnxruntime::hosting::GetRequestFromJson(input_json, request);
+  onnxruntime::server::PredictRequest request;
+  protobufutil::Status status = onnxruntime::server::GetRequestFromJson(input_json, request);
 
   EXPECT_EQ(protobufutil::error::OK, status.error_code());
 }
 
 TEST(JsonDeserializationTests, WithUnknownField) {
   std::string input_json = R"({"foo": "bar","inputs":{"Input3":{"dims":["1","1","28","28"],"dataType":1,"rawData":"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"}},"outputFilter":["Plus214_Output_0"]})";
-  onnxruntime::hosting::PredictRequest request;
-  protobufutil::Status status = onnxruntime::hosting::GetRequestFromJson(input_json, request);
+  onnxruntime::server::PredictRequest request;
+  protobufutil::Status status = onnxruntime::server::GetRequestFromJson(input_json, request);
 
   EXPECT_EQ(protobufutil::error::OK, status.error_code());
 }
 
 TEST(JsonDeserializationTests, InvalidData) {
   std::string input_json = R"({"inputs":{"Input3":{"dims":["1","1","28","28"],"dataType":1,"rawData":"hello"}},"outputFilter":["Plus214_Output_0"]})";
-  onnxruntime::hosting::PredictRequest request;
-  protobufutil::Status status = onnxruntime::hosting::GetRequestFromJson(input_json, request);
+  onnxruntime::server::PredictRequest request;
+  protobufutil::Status status = onnxruntime::server::GetRequestFromJson(input_json, request);
 
   EXPECT_EQ(protobufutil::error::INVALID_ARGUMENT, status.error_code());
   EXPECT_EQ("inputs[0].value.raw_data: invalid value \"hello\" for type TYPE_BYTES", status.error_message());
@@ -41,8 +41,8 @@ TEST(JsonDeserializationTests, InvalidData) {
 
 TEST(JsonDeserializationTests, InvalidJson) {
   std::string input_json = R"({inputs":{"Input3":{"dims":["1","1","28","28"],"dataType":1,"rawData":"hello"}},"outputFilter":["Plus214_Output_0"]})";
-  onnxruntime::hosting::PredictRequest request;
-  protobufutil::Status status = onnxruntime::hosting::GetRequestFromJson(input_json, request);
+  onnxruntime::server::PredictRequest request;
+  protobufutil::Status status = onnxruntime::server::GetRequestFromJson(input_json, request);
 
   EXPECT_EQ(protobufutil::error::INVALID_ARGUMENT, status.error_code());
   std::string errmsg = status.error_message();
@@ -50,9 +50,9 @@ TEST(JsonDeserializationTests, InvalidJson) {
 }
 
 TEST(JsonSerializationTests, HappyPath) {
-  std::string test_data = "testdata/hosting/response_0.pb";
+  std::string test_data = "testdata/server/response_0.pb";
   std::string expected_json_string = R"({"outputs":{"Plus214_Output_0":{"dims":["1","10"],"dataType":1,"rawData":"4+pzRFWuGsSMdM1F2gEnRFdRZcRZ9NDEURj0xBIzdsJOS0LEA/GzxA=="}}})";
-  onnxruntime::hosting::PredictResponse response;
+  onnxruntime::server::PredictResponse response;
   std::string json_string;
 
   std::ifstream ifs(test_data, std::ios_base::in | std::ios_base::binary);
@@ -62,7 +62,7 @@ TEST(JsonSerializationTests, HappyPath) {
   ifs.close();
   EXPECT_TRUE(succeeded) << test_data << " is invalid" << std::endl;
 
-  protobufutil::Status status = onnxruntime::hosting::GenerateResponseInJson(response, json_string);
+  protobufutil::Status status = onnxruntime::server::GenerateResponseInJson(response, json_string);
 
   EXPECT_EQ(protobufutil::error::OK, status.error_code());
   EXPECT_EQ(expected_json_string, json_string);
@@ -124,5 +124,5 @@ TEST(JsonErrorMessageTests, MessageWithManyCarriageCharacters) {
 }
 
 }  // namespace test
-}  // namespace hosting
+}  // namespace server
 }  // namespace onnxruntime
\ No newline at end of file
diff --git a/onnxruntime/test/hosting/unit_tests/server_configuration_test.cc b/onnxruntime/test/server/unit_tests/server_configuration_test.cc
similarity index 88%
rename from onnxruntime/test/hosting/unit_tests/server_configuration_test.cc
rename to onnxruntime/test/server/unit_tests/server_configuration_test.cc
index bc2210374902b..77a979e27f593 100644
--- a/onnxruntime/test/hosting/unit_tests/server_configuration_test.cc
+++ b/onnxruntime/test/server/unit_tests/server_configuration_test.cc
@@ -4,10 +4,10 @@
 #include "gtest/gtest.h"
 
 #include "gmock/gmock.h"
-#include "hosting/server_configuration.h"
+#include "server/server_configuration.h"
 
 namespace onnxruntime {
-namespace hosting {
+namespace server {
 namespace test {
 
 TEST(ConfigParsingTests, AllArgs) {
@@ -19,7 +19,7 @@ TEST(ConfigParsingTests, AllArgs) {
       const_cast<char*>("--num_http_threads"), const_cast<char*>("1"),
       const_cast<char*>("--log_level"), const_cast<char*>("info")};
 
-  onnxruntime::hosting::ServerConfiguration config{};
+  onnxruntime::server::ServerConfiguration config{};
   Result res = config.ParseInput(11, test_argv);
   EXPECT_EQ(res, Result::ContinueSuccess);
   EXPECT_EQ(config.model_path, "testdata/mul_1.pb");
@@ -35,7 +35,7 @@ TEST(ConfigParsingTests, Defaults) {
       const_cast<char*>("--model"), const_cast<char*>("testdata/mul_1.pb"),
       const_cast<char*>("--num_http_threads"), const_cast<char*>("3")};
 
-  onnxruntime::hosting::ServerConfiguration config{};
+  onnxruntime::server::ServerConfiguration config{};
   Result res = config.ParseInput(5, test_argv);
   EXPECT_EQ(res, Result::ContinueSuccess);
   EXPECT_EQ(config.model_path, "testdata/mul_1.pb");
@@ -50,7 +50,7 @@ TEST(ConfigParsingTests, Help) {
       const_cast<char*>("/path/to/binary"),
       const_cast<char*>("--help")};
 
-  onnxruntime::hosting::ServerConfiguration config{};
+  onnxruntime::server::ServerConfiguration config{};
   auto res = config.ParseInput(2, test_argv);
   EXPECT_EQ(res, Result::ExitSuccess);
 }
@@ -60,7 +60,7 @@ TEST(ConfigParsingTests, NoModelArg) {
       const_cast<char*>("/path/to/binary"),
       const_cast<char*>("--num_http_threads"), const_cast<char*>("3")};
 
-  onnxruntime::hosting::ServerConfiguration config{};
+  onnxruntime::server::ServerConfiguration config{};
   Result res = config.ParseInput(3, test_argv);
   EXPECT_EQ(res, Result::ExitFailure);
 }
@@ -73,7 +73,7 @@ TEST(ConfigParsingTests, ModelNotFound) {
       const_cast<char*>("--http_port"), const_cast<char*>("80"),
       const_cast<char*>("--num_http_threads"), const_cast<char*>("1")};
 
-  onnxruntime::hosting::ServerConfiguration config{};
+  onnxruntime::server::ServerConfiguration config{};
   Result res = config.ParseInput(9, test_argv);
   EXPECT_EQ(res, Result::ExitFailure);
 }
@@ -87,11 +87,11 @@ TEST(ConfigParsingTests, WrongLoggingLevel) {
       const_cast<char*>("--http_port"), const_cast<char*>("80"),
       const_cast<char*>("--num_http_threads"), const_cast<char*>("1")};
 
-  onnxruntime::hosting::ServerConfiguration config{};
+  onnxruntime::server::ServerConfiguration config{};
   Result res = config.ParseInput(11, test_argv);
   EXPECT_EQ(res, Result::ExitFailure);
 }
 
 }  // namespace test
-}  // namespace hosting
+}  // namespace server
 }  // namespace onnxruntime
\ No newline at end of file
diff --git a/onnxruntime/test/hosting/unit_tests/test_main.cc b/onnxruntime/test/server/unit_tests/test_main.cc
similarity index 100%
rename from onnxruntime/test/hosting/unit_tests/test_main.cc
rename to onnxruntime/test/server/unit_tests/test_main.cc
diff --git a/onnxruntime/test/hosting/unit_tests/util_tests.cc b/onnxruntime/test/server/unit_tests/util_tests.cc
similarity index 97%
rename from onnxruntime/test/hosting/unit_tests/util_tests.cc
rename to onnxruntime/test/server/unit_tests/util_tests.cc
index 9ed7aaa8f9334..c16f87c4a712b 100644
--- a/onnxruntime/test/hosting/unit_tests/util_tests.cc
+++ b/onnxruntime/test/server/unit_tests/util_tests.cc
@@ -3,11 +3,11 @@
 
 #include <google/protobuf/stubs/status.h>
 #include "gtest/gtest.h"
-#include "hosting/http/core/context.h"
-#include "hosting/http/util.h"
+#include "server/http/core/context.h"
+#include "server/http/util.h"
 
 namespace onnxruntime {
-namespace hosting {
+namespace server {
 namespace test {
 
 namespace protobufutil = google::protobuf::util;
@@ -117,5 +117,5 @@ TEST(ContentTypeTests, ContentTypeMissing) {
 }
 
 }  // namespace test
-}  // namespace hosting
+}  // namespace server
 }  // namespace onnxruntime
\ No newline at end of file
diff --git a/onnxruntime/test/testdata/hosting/mnist.onnx b/onnxruntime/test/testdata/hosting/mnist.onnx
deleted file mode 100644
index fc1a3f733c6e6..0000000000000
Binary files a/onnxruntime/test/testdata/hosting/mnist.onnx and /dev/null differ
diff --git a/onnxruntime/test/testdata/hosting/mnist_test_data_set_0_input.json b/onnxruntime/test/testdata/server/mnist_test_data_set_0_input.json
similarity index 100%
rename from onnxruntime/test/testdata/hosting/mnist_test_data_set_0_input.json
rename to onnxruntime/test/testdata/server/mnist_test_data_set_0_input.json
diff --git a/onnxruntime/test/testdata/hosting/mnist_test_data_set_0_input.pb b/onnxruntime/test/testdata/server/mnist_test_data_set_0_input.pb
similarity index 100%
rename from onnxruntime/test/testdata/hosting/mnist_test_data_set_0_input.pb
rename to onnxruntime/test/testdata/server/mnist_test_data_set_0_input.pb
diff --git a/onnxruntime/test/testdata/hosting/mnist_test_data_set_0_output.json b/onnxruntime/test/testdata/server/mnist_test_data_set_0_output.json
similarity index 100%
rename from onnxruntime/test/testdata/hosting/mnist_test_data_set_0_output.json
rename to onnxruntime/test/testdata/server/mnist_test_data_set_0_output.json
diff --git a/onnxruntime/test/testdata/hosting/mnist_test_data_set_0_output.pb b/onnxruntime/test/testdata/server/mnist_test_data_set_0_output.pb
similarity index 100%
rename from onnxruntime/test/testdata/hosting/mnist_test_data_set_0_output.pb
rename to onnxruntime/test/testdata/server/mnist_test_data_set_0_output.pb
diff --git a/onnxruntime/test/testdata/hosting/request_0.pb b/onnxruntime/test/testdata/server/request_0.pb
similarity index 100%
rename from onnxruntime/test/testdata/hosting/request_0.pb
rename to onnxruntime/test/testdata/server/request_0.pb
diff --git a/onnxruntime/test/testdata/hosting/response_0.pb b/onnxruntime/test/testdata/server/response_0.pb
similarity index 100%
rename from onnxruntime/test/testdata/hosting/response_0.pb
rename to onnxruntime/test/testdata/server/response_0.pb
diff --git a/tools/ci_build/build.py b/tools/ci_build/build.py
index 0d0645ff77853..e3f2edd186a33 100755
--- a/tools/ci_build/build.py
+++ b/tools/ci_build/build.py
@@ -95,9 +95,9 @@ def parse_arguments():
     # Build a shared lib
     parser.add_argument("--build_shared_lib", action='store_true', help="Build a shared library for the ONNXRuntime.")
 
-    # Build ONNX hosting
-    parser.add_argument("--build_hosting", action='store_true', help="Build hosting application for the ONNXRuntime.")
-    parser.add_argument("--enable_hosting_tests", action='store_true', help="Run hosting application tests.")
+    # Build ONNX Runtime server
+    parser.add_argument("--build_server", action='store_true', help="Build server application for the ONNXRuntime.")
+    parser.add_argument("--enable_server_tests", action='store_true', help="Run server application tests.")
 
     # Build options
     parser.add_argument("--cmake_extra_defines", nargs="+",
@@ -328,10 +328,10 @@ def generate_build_tree(cmake_path, source_dir, build_dir, cuda_home, cudnn_home
                  "-Donnxruntime_TENSORRT_HOME=" + (tensorrt_home if args.use_tensorrt else ""),
                   # By default - we currently support only cross compiling for ARM/ARM64 (no native compilation supported through this script)
                  "-Donnxruntime_CROSS_COMPILING=" + ("ON" if args.arm64 or args.arm else "OFF"),
-                 "-Donnxruntime_BUILD_HOSTING=" + ("ON" if args.build_hosting else "OFF"),
+                 "-Donnxruntime_BUILD_SERVER=" + ("ON" if args.build_server else "OFF"),
                  "-Donnxruntime_BUILD_x86=" + ("ON" if args.x86 else "OFF"),
                   # nGraph and TensorRT providers currently only supports full_protobuf option.
-                 "-Donnxruntime_USE_FULL_PROTOBUF=" + ("ON" if args.use_full_protobuf or args.use_ngraph or args.use_tensorrt or args.build_hosting else "OFF"),
+                 "-Donnxruntime_USE_FULL_PROTOBUF=" + ("ON" if args.use_full_protobuf or args.use_ngraph or args.use_tensorrt or args.build_server else "OFF"),
                  "-Donnxruntime_DISABLE_CONTRIB_OPS=" + ("ON" if args.disable_contrib_ops else "OFF"),
                  "-Donnxruntime_MSVC_STATIC_RUNTIME=" + ("ON" if args.enable_msvc_static_runtime else "OFF"),
                  ]
@@ -571,14 +571,14 @@ def run_onnx_tests(build_dir, configs, onnx_test_data_dir, provider, enable_para
           run_subprocess([exe,'-x'] + cmd, cwd=cwd)
 
 
-def run_hosting_tests(build_dir, configs):
+def run_server_tests(build_dir, configs):
     run_subprocess([sys.executable, '-m', 'pip', 'install', '--trusted-host', 'files.pythonhosted.org', 'requests'])
     for config in configs:
         config_build_dir = get_config_build_dir(build_dir, config)
-        hosting_app_path = os.path.join(config_build_dir, 'onnxruntime_hosting')
-        hosting_test_folder = os.path.join(config_build_dir, 'hosting_test')
-        hosting_test_data_folder = os.path.join(os.path.join(config_build_dir, 'testdata'), 'hosting')
-        run_subprocess([sys.executable, 'test_main.py', hosting_app_path, hosting_test_data_folder, hosting_test_data_folder], cwd=hosting_test_folder, dll_path=None)
+        server_app_path = os.path.join(config_build_dir, 'onnxruntime_server')
+        server_test_folder = os.path.join(config_build_dir, 'server_test')
+        server_test_data_folder = os.path.join(os.path.join(config_build_dir, 'testdata'), 'server')
+        run_subprocess([sys.executable, 'test_main.py', server_app_path, server_test_data_folder, server_test_data_folder], cwd=server_test_folder, dll_path=None)
 
 def build_python_wheel(source_dir, build_dir, configs, use_cuda, use_ngraph, use_tensorrt, nightly_build = False):
     for config in configs:
@@ -781,8 +781,8 @@ def main():
               if args.use_mkldnn:
                 run_onnx_tests(build_dir, configs, onnx_test_data_dir, 'mkldnn', True, 1)
 
-    if args.build_hosting and args.enable_hosting_tests:
-        run_hosting_tests(build_dir, configs)
+    if args.build_server and args.enable_server_tests:
+        run_server_tests(build_dir, configs)
 
     if args.build:
         if args.build_wheel:
diff --git a/tools/ci_build/github/azure-pipelines/vienna-linux-ci-pipeline.yml b/tools/ci_build/github/azure-pipelines/vienna-linux-ci-pipeline.yml
index 572cce584ece5..16d0fe959958a 100644
--- a/tools/ci_build/github/azure-pipelines/vienna-linux-ci-pipeline.yml
+++ b/tools/ci_build/github/azure-pipelines/vienna-linux-ci-pipeline.yml
@@ -3,7 +3,7 @@ jobs:
   pool: Hosted Ubuntu 1604
   steps:
     - template: templates/set-test-data-variables-step.yml
-    - script: 'tools/ci_build/github/linux/hosting_run_dockerbuild.sh -o ubuntu16.04 -d cpu -r $(Build.BinariesDirectory) -k $(acr.key) -x "--config Debug --build_hosting --use_openmp --use_full_protobuf --enable_hosting_tests"'
+    - script: 'tools/ci_build/github/linux/server_run_dockerbuild.sh -o ubuntu16.04 -d cpu -r $(Build.BinariesDirectory) -k $(acr.key) -x "--config Debug --build_server --use_openmp --use_full_protobuf --enable_server_tests"'
       displayName: 'Debug Build'
     - task: ms.vss-governance-buildtask.governance-build-task-component-detection.ComponentGovernanceComponentDetection@0
       displayName: 'Component Detection'
@@ -12,7 +12,7 @@ jobs:
   pool: Hosted Ubuntu 1604
   steps:
     - template: templates/set-test-data-variables-step.yml
-    - script: 'tools/ci_build/github/linux/hosting_run_dockerbuild.sh -o ubuntu16.04 -d cpu -r $(Build.BinariesDirectory) -k $(acr.key) -x "--config Release --build_hosting --use_openmp --use_full_protobuf --enable_hosting_tests"'
+    - script: 'tools/ci_build/github/linux/server_run_dockerbuild.sh -o ubuntu16.04 -d cpu -r $(Build.BinariesDirectory) -k $(acr.key) -x "--config Release --build_server --use_openmp --use_full_protobuf --enable_server_tests"'
       displayName: 'Release Build'
     - task: ms.vss-governance-buildtask.governance-build-task-component-detection.ComponentGovernanceComponentDetection@0
       displayName: 'Component Detection'
diff --git a/tools/ci_build/github/linux/hosting_run_build.sh b/tools/ci_build/github/linux/server_run_build.sh
similarity index 100%
rename from tools/ci_build/github/linux/hosting_run_build.sh
rename to tools/ci_build/github/linux/server_run_build.sh
diff --git a/tools/ci_build/github/linux/hosting_run_dockerbuild.sh b/tools/ci_build/github/linux/server_run_dockerbuild.sh
similarity index 93%
rename from tools/ci_build/github/linux/hosting_run_dockerbuild.sh
rename to tools/ci_build/github/linux/server_run_dockerbuild.sh
index 7e0ea192d946b..87869b1df811d 100755
--- a/tools/ci_build/github/linux/hosting_run_dockerbuild.sh
+++ b/tools/ci_build/github/linux/server_run_dockerbuild.sh
@@ -55,7 +55,7 @@ fi
         --volume "$BUILD_DIR:/home/onnxruntimedev" \
         --volume "$HOME/.cache/onnxruntime:/home/onnxruntimedev/.cache/onnxruntime" \
         "onnxruntime-$IMAGE" \
-        /bin/bash /onnxruntime_src/tools/ci_build/github/linux/hosting_run_build.sh \
+        /bin/bash /onnxruntime_src/tools/ci_build/github/linux/server_run_build.sh \
          -d $BUILD_DEVICE -x "$BUILD_EXTR_PAR" &
 else
     docker rm -f "onnxruntime-$BUILD_DEVICE" || true
@@ -66,7 +66,7 @@ else
         --volume "$BUILD_DIR:/home/onnxruntimedev" \
         --volume "$HOME/.cache/onnxruntime:/home/onnxruntimedev/.cache/onnxruntime" \
         "onnxruntime-$IMAGE" \
-        /bin/bash /onnxruntime_src/tools/ci_build/github/linux/hosting_run_build.sh \
+        /bin/bash /onnxruntime_src/tools/ci_build/github/linux/server_run_build.sh \
         -d $BUILD_DEVICE -x "$BUILD_EXTR_PAR" &
 fi
 wait -n