From 2bb9020d7da96818f411aef0b0d2bf53c2663a86 Mon Sep 17 00:00:00 2001 From: Pavithra Vijayakrishnan <160681768+pvijayakrish@users.noreply.github.com> Date: Thu, 25 Jul 2024 10:36:03 -0700 Subject: [PATCH 1/3] Update NGC versions post-24.07 release (#7469) --- Dockerfile.sdk | 2 +- Dockerfile.win10.min | 20 ++++----- README.md | 17 ++++---- TRITON_VERSION | 2 +- build.py | 41 ++++++++++++------- deploy/aws/values.yaml | 6 +-- deploy/fleetcommand/Chart.yaml | 4 +- deploy/fleetcommand/values.yaml | 8 ++-- deploy/gcp/values.yaml | 6 +-- .../perf-analyzer-script/triton_client.yaml | 4 +- .../server-deployer/build_and_push.sh | 8 ++-- .../server-deployer/chart/triton/Chart.yaml | 6 +-- .../server-deployer/chart/triton/values.yaml | 8 ++-- .../server-deployer/data-test/schema.yaml | 4 +- .../server-deployer/schema.yaml | 6 +-- .../gke-marketplace-app/trt-engine/README.md | 8 ++-- deploy/k8s-onprem/values.yaml | 6 +-- deploy/oci/values.yaml | 6 +-- docs/customization_guide/build.md | 8 ++-- docs/customization_guide/compose.md | 20 ++++----- docs/customization_guide/test.md | 4 +- docs/generate_docs.py | 4 +- docs/user_guide/custom_operations.md | 8 ++-- docs/user_guide/metrics.md | 2 +- docs/user_guide/performance_tuning.md | 6 +-- qa/L0_tftrt_optimization/test.sh | 2 - qa/L0_warmup/test.sh | 4 -- qa/common/gen_jetson_trt_models | 4 +- qa/common/gen_qa_custom_ops | 4 +- qa/common/gen_qa_model_repository | 2 +- 30 files changed, 119 insertions(+), 111 deletions(-) diff --git a/Dockerfile.sdk b/Dockerfile.sdk index 9e83ecca47..e92b4bcb89 100644 --- a/Dockerfile.sdk +++ b/Dockerfile.sdk @@ -29,7 +29,7 @@ # # Base image on the minimum Triton container -ARG BASE_IMAGE=nvcr.io/nvidia/tritonserver:24.06-py3-min +ARG BASE_IMAGE=nvcr.io/nvidia/tritonserver:24.07-py3-min ARG TRITON_CLIENT_REPO_SUBDIR=clientrepo ARG TRITON_COMMON_REPO_TAG=main diff --git a/Dockerfile.win10.min b/Dockerfile.win10.min index 7d954d62de..0a554fbcf4 100644 --- a/Dockerfile.win10.min +++ b/Dockerfile.win10.min @@ -1,4 +1,4 @@ -# Copyright 2021-2022, NVIDIA CORPORATION & AFFILIATES. All rights reserved. +# Copyright 2021-2024, NVIDIA CORPORATION & AFFILIATES. All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions @@ -37,9 +37,9 @@ RUN choco install unzip -y # # Installing TensorRT # -ARG TENSORRT_VERSION=10.0.1.6 -ARG TENSORRT_ZIP="TensorRT-${TENSORRT_VERSION}.Windows10.x86_64.cuda-12.4.zip" -ARG TENSORRT_SOURCE=https://developer.nvidia.com/downloads/compute/machine-learning/tensorrt/10.0.1/zip/TensorRT-10.0.1.6.Windows10.win10.cuda-12.4.zip +ARG TENSORRT_VERSION=10.2.0.19 +ARG TENSORRT_ZIP="TensorRT-${TENSORRT_VERSION}.Windows10.x86_64.cuda-12.5.zip" +ARG TENSORRT_SOURCE=https://developer.nvidia.com/downloads/compute/machine-learning/tensorrt/10.2.0/zip/TensorRT-10.2.0.19.Windows10.x86_64.cuda-12.5.zip # COPY ${TENSORRT_ZIP} /tmp/${TENSORRT_ZIP} ADD ${TENSORRT_SOURCE} /tmp/${TENSORRT_ZIP} RUN unzip /tmp/%TENSORRT_ZIP% @@ -51,9 +51,9 @@ LABEL TENSORRT_VERSION="${TENSORRT_VERSION}" # # Installing cuDNN # -ARG CUDNN_VERSION=9.1.0.70 +ARG CUDNN_VERSION=9.2.1.18 ARG CUDNN_ZIP=cudnn-windows-x86_64-${CUDNN_VERSION}_cuda12-archive.zip -ARG CUDNN_SOURCE=https://developer.download.nvidia.com/compute/cudnn/redist/cudnn/windows-x86_64/cudnn-windows-x86_64-9.1.0.70_cuda12-archive.zip +ARG CUDNN_SOURCE=https://developer.download.nvidia.com/compute/cudnn/redist/cudnn/windows-x86_64/cudnn-windows-x86_64-9.2.1.18_cuda12-archive.zip ADD ${CUDNN_SOURCE} /tmp/${CUDNN_ZIP} RUN unzip /tmp/%CUDNN_ZIP% RUN move cudnn-* cudnn @@ -88,7 +88,7 @@ LABEL PYTHON_VERSION=${PYTHON_VERSION} # # Installing CMake # -ARG CMAKE_VERSION=3.29.3 +ARG CMAKE_VERSION=3.30.0 RUN pip install cmake==%CMAKE_VERSION% ENV CMAKE_TOOLCHAIN_FILE /vcpkg/scripts/buildsystems/vcpkg.cmake @@ -150,7 +150,7 @@ WORKDIR / # ARG CUDA_MAJOR=12 ARG CUDA_MINOR=5 -ARG CUDA_PATCH=0 +ARG CUDA_PATCH=1 ARG CUDA_VERSION=${CUDA_MAJOR}.${CUDA_MINOR}.${CUDA_PATCH} ARG CUDA_PACKAGES="nvcc_${CUDA_MAJOR}.${CUDA_MINOR} \ cudart_${CUDA_MAJOR}.${CUDA_MINOR} \ @@ -175,7 +175,7 @@ RUN copy "%CUDA_INSTALL_ROOT_WP%\extras\visual_studio_integration\MSBuildExtensi RUN setx PATH "%CUDA_INSTALL_ROOT_WP%\bin;%PATH%" -ARG CUDNN_VERSION=9.1.0.70 +ARG CUDNN_VERSION=9.2.1.18 ENV CUDNN_VERSION ${CUDNN_VERSION} COPY --from=dependency_base /cudnn /cudnn RUN copy cudnn\bin\cudnn*.dll "%CUDA_INSTALL_ROOT_WP%\bin\." @@ -183,7 +183,7 @@ RUN copy cudnn\lib\x64\cudnn*.lib "%CUDA_INSTALL_ROOT_WP%\lib\x64\." RUN copy cudnn\include\cudnn*.h "%CUDA_INSTALL_ROOT_WP%\include\." LABEL CUDNN_VERSION="${CUDNN_VERSION}" -ARG TENSORRT_VERSION=10.0.1.6 +ARG TENSORRT_VERSION=10.2.0.19 ENV TRT_VERSION ${TENSORRT_VERSION} COPY --from=dependency_base /TensorRT /TensorRT RUN setx PATH "c:\TensorRT\lib;%PATH%" diff --git a/README.md b/README.md index 38b4759c48..17628b4f03 100644 --- a/README.md +++ b/README.md @@ -1,5 +1,5 @@