From 3cf9dec1976bfec3fc1b36bec7eb5c6548e34227 Mon Sep 17 00:00:00 2001 From: Silva Alejandro Ismael Date: Mon, 17 Jun 2024 15:58:10 -0300 Subject: [PATCH 1/3] catalyst-latam: fix GPUs for plnc hub --- .../catalystproject-latam/plnc.values.yaml | 19 +++++++++++-------- 1 file changed, 11 insertions(+), 8 deletions(-) diff --git a/config/clusters/catalystproject-latam/plnc.values.yaml b/config/clusters/catalystproject-latam/plnc.values.yaml index 73828812fb..fade119c12 100644 --- a/config/clusters/catalystproject-latam/plnc.values.yaml +++ b/config/clusters/catalystproject-latam/plnc.values.yaml @@ -35,9 +35,9 @@ jupyterhub: image: quay.io/jupyter/scipy-notebook:2024-03-04 default_url: /lab profile_options: &profile_options - resource_allocation: &resource_allocation + resource_allocation: display_name: Resource Allocation - choices: + choices: &resource_choices mem_0_7: display_name: Up to 2G of RAM and 1 CPU kubespawner_override: @@ -70,16 +70,16 @@ jupyterhub: working_dir: /home/rstudio # Ensures container working dir is homedir profile_options: *profile_options - - display_name: Pangeo Tensorflow ML Notebook + - display_name: Tensorflow ML Notebook description: Dedicated node with a NVIDIA Tesla T4 slug: "tensorflow" kubespawner_override: - image: "quay.io/pangeo/ml-notebook:2024.05.07" + image: "quay.io/jupyter/tensorflow-notebook:cuda-2024-06-17" default_url: /lab profile_options: &profile_gpu_options resource_allocation: display_name: Resource Allocation - choices: + choices: &resource_choices_gpu mem_26_0: display_name: 1 GPU, 21G of RAM and 3 CPUs default: true @@ -106,11 +106,11 @@ jupyterhub: extra_resource_limits: nvidia.com/gpu: "4" - - display_name: Pangeo PyTorch ML Notebook + - display_name: PyTorch ML Notebook description: Dedicated node with a NVIDIA Tesla T4 slug: "pytorch" kubespawner_override: - image: "quay.io/pangeo/pytorch-notebook:2024.05.07" + image: "quay.io/jupyter/pytorch-notebook:cuda12-2024-06-17" default_url: /lab profile_options: *profile_gpu_options @@ -128,4 +128,7 @@ jupyterhub: kubespawner_override: image: "{value}" choices: {} - resource_allocation: *resource_allocation + resource_allocation: + display_name: Resource Allocation + choices: + <<: [*resource_choices, *resource_choices_gpu] From 5ffbd369d47d7bd440add9d342e3d73c205ba225 Mon Sep 17 00:00:00 2001 From: Silva Alejandro Ismael Date: Fri, 21 Jun 2024 12:35:23 -0300 Subject: [PATCH 2/3] catalyst-latam: simplified profileList of plnc hub --- .../catalystproject-latam/plnc.values.yaml | 96 ++++++++++--------- 1 file changed, 50 insertions(+), 46 deletions(-) diff --git a/config/clusters/catalystproject-latam/plnc.values.yaml b/config/clusters/catalystproject-latam/plnc.values.yaml index fade119c12..0861f45805 100644 --- a/config/clusters/catalystproject-latam/plnc.values.yaml +++ b/config/clusters/catalystproject-latam/plnc.values.yaml @@ -27,15 +27,14 @@ jupyterhub: singleuser: profileList: - - display_name: Jupyter SciPy Notebook - description: Python environment - slug: jupyter + - display_name: "CPU only" + description: "Start a container limited to a chosen share of capacity" + slug: cpu default: true kubespawner_override: - image: quay.io/jupyter/scipy-notebook:2024-03-04 default_url: /lab - profile_options: &profile_options - resource_allocation: + profile_options: + requests: display_name: Resource Allocation choices: &resource_choices mem_0_7: @@ -59,27 +58,38 @@ jupyterhub: mem_limit: 24G cpu_guarantee: 1.5 cpu_limit: 3 + image: + display_name: Image + unlisted_choice: + enabled: True + display_name: "Custom image" + validation_regex: "^.+:.+$" + validation_message: "Must be a publicly available docker image, of form :" + kubespawner_override: + image: "{value}" + choices: + rstudio: + display_name: Rocker Geospatial with RStudio + default: true + slug: "rstudio" + kubespawner_override: + image: "rocker/binder:4.3" + image_pull_policy: Always + default_url: /rstudio + working_dir: /home/rstudio # Ensures container working dir is homedir + jupyter: + display_name: Jupyter SciPy Notebook + slug: "jupyter" + kubespawner_override: + image: "quay.io/jupyter/scipy-notebook:2024-03-04" + default_url: /lab - - display_name: Rocker Geospatial with RStudio - description: R environment - slug: rocker - kubespawner_override: - image: rocker/binder:4.3 - image_pull_policy: Always - default_url: /rstudio - working_dir: /home/rstudio # Ensures container working dir is homedir - profile_options: *profile_options - - - display_name: Tensorflow ML Notebook - description: Dedicated node with a NVIDIA Tesla T4 - slug: "tensorflow" - kubespawner_override: - image: "quay.io/jupyter/tensorflow-notebook:cuda-2024-06-17" - default_url: /lab - profile_options: &profile_gpu_options - resource_allocation: + - display_name: "With GPUs" + description: "Start a container on a dedicated node with a NVIDIA Tesla T4" + profile_options: + requests: display_name: Resource Allocation - choices: &resource_choices_gpu + choices: mem_26_0: display_name: 1 GPU, 21G of RAM and 3 CPUs default: true @@ -87,8 +97,6 @@ jupyterhub: mem_guarantee: 21G mem_limit: null cpu_guarantee: 3 - environment: - NVIDIA_DRIVER_CAPABILITIES: compute,utility node_selector: node.kubernetes.io/instance-type: n1-highmem-4 extra_resource_limits: @@ -99,25 +107,10 @@ jupyterhub: mem_guarantee: 92G mem_limit: null cpu_guarantee: 15 - environment: - NVIDIA_DRIVER_CAPABILITIES: compute,utility node_selector: node.kubernetes.io/instance-type: n1-highmem-16 extra_resource_limits: nvidia.com/gpu: "4" - - - display_name: PyTorch ML Notebook - description: Dedicated node with a NVIDIA Tesla T4 - slug: "pytorch" - kubespawner_override: - image: "quay.io/jupyter/pytorch-notebook:cuda12-2024-06-17" - default_url: /lab - profile_options: *profile_gpu_options - - - display_name: Bring your own image - description: Specify your own docker image (must have python and jupyterhub installed in it) - slug: custom - profile_options: image: display_name: Image unlisted_choice: @@ -127,8 +120,19 @@ jupyterhub: validation_message: "Must be a publicly available docker image, of form :" kubespawner_override: image: "{value}" - choices: {} - resource_allocation: - display_name: Resource Allocation choices: - <<: [*resource_choices, *resource_choices_gpu] + tensorflow: + display_name: Tensorflow ML Notebook + slug: "tensorflow" + kubespawner_override: + image: "quay.io/jupyter/tensorflow-notebook:cuda-2024-06-17" + pytorch: + display_name: PyTorch ML Notebook + default: true + slug: "pytorch" + kubespawner_override: + image: "quay.io/jupyter/pytorch-notebook:cuda12-2024-06-17" + kubespawner_override: + default_url: /lab + environment: + NVIDIA_DRIVER_CAPABILITIES: compute,utility From 519413cd178997d8397aaacabd2ef0bb19895eae Mon Sep 17 00:00:00 2001 From: Erik Sundell Date: Mon, 1 Jul 2024 16:55:06 +0200 Subject: [PATCH 3/3] catalyst-latam, plnc: update image tags for use with gpu with recent fixes --- config/clusters/catalystproject-latam/plnc.values.yaml | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/config/clusters/catalystproject-latam/plnc.values.yaml b/config/clusters/catalystproject-latam/plnc.values.yaml index 0861f45805..5755782f69 100644 --- a/config/clusters/catalystproject-latam/plnc.values.yaml +++ b/config/clusters/catalystproject-latam/plnc.values.yaml @@ -125,13 +125,13 @@ jupyterhub: display_name: Tensorflow ML Notebook slug: "tensorflow" kubespawner_override: - image: "quay.io/jupyter/tensorflow-notebook:cuda-2024-06-17" + image: "quay.io/jupyter/tensorflow-notebook:cuda-2024-07-01" pytorch: display_name: PyTorch ML Notebook default: true slug: "pytorch" kubespawner_override: - image: "quay.io/jupyter/pytorch-notebook:cuda12-2024-06-17" + image: "quay.io/jupyter/pytorch-notebook:cuda12-2024-07-01" kubespawner_override: default_url: /lab environment: