From 874da1acb531d2cf19ec96ff1e10b5140d26e2ae Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Rapha=C3=ABl=20Bournhonesque?= Date: Thu, 12 Dec 2024 16:39:14 +0100 Subject: [PATCH] chore: increase allowed memory for Triton --- docker/ml-gpu.yml | 2 +- docker/ml.yml | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/docker/ml-gpu.yml b/docker/ml-gpu.yml index 3f90d1d5da..5a3bfbda5e 100644 --- a/docker/ml-gpu.yml +++ b/docker/ml-gpu.yml @@ -22,7 +22,7 @@ services: # https://github.com/triton-inference-server/server/blob/main/docs/user_guide/model_management.md # for more information entrypoint: "/opt/nvidia/nvidia_entrypoint.sh tritonserver --model-repository=/models --model-control-mode=explicit --load-model=*" - mem_limit: 15g + mem_limit: 20g runtime: nvidia deploy: resources: diff --git a/docker/ml.yml b/docker/ml.yml index b7dca98531..c348138b70 100644 --- a/docker/ml.yml +++ b/docker/ml.yml @@ -14,7 +14,7 @@ services: # https://github.com/triton-inference-server/server/blob/main/docs/user_guide/model_management.md # for more information entrypoint: "tritonserver --model-repository=/models --model-control-mode=explicit --load-model=*" - mem_limit: 15g + mem_limit: 20g fasttext: restart: $RESTART_POLICY