--- title: Amazon ECS now supports GPU sharing with NVIDIA... description: Amazon ECS now supports GPU sharing through NVIDIA Multi-Process Service, letting multiple ECS tasks run concurrently on a single GPU. This helps increase GPU... canonical: https://daily.dev/posts/amazon-ecs-now-supports-gpu-sharing-with-nvidia-multi-process-service-mps--dbsrt4pms twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Amazon ECS now supports GPU sharing with NVIDIA Multi-Process Service (MPS) | daily.dev og:description: Amazon ECS now supports GPU sharing through NVIDIA Multi-Process Service, letting multiple ECS tasks run concurrently on a single GPU. This helps increase GPU... og:url: https://daily.dev/posts/amazon-ecs-now-supports-gpu-sharing-with-nvidia-multi-process-service-mps--dbsrt4pms og:image: https://api.daily.dev/og/posts/dbSrT4PMs.png og:image:alt: Amazon ECS now supports GPU sharing with NVIDIA Multi-Process Service (MPS) og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Amazon ECS now supports GPU sharing with NVIDIA Multi-Process Service (MPS) **[AWS](https://daily.dev/sources/aws)** · 2 min read · 0 upvotes · 0 comments ## Summary Amazon ECS now supports GPU sharing through NVIDIA Multi-Process Service, letting multiple ECS tasks run concurrently on a single GPU. This helps increase GPU utilization and cut infrastructure costs for workloads that only need part of a GPU, like small-model inference, AV/ADAS simulations, embedding generation, and multi-model serving. You define GPU memory per container and can set a max GPU compute share, and ECS places tasks on GPUs with available capacity. On ECS Managed Instances, ECS monitors the MPS control daemon and automatically replaces unhealthy instances. The feature works on ECS Managed Instances and ECS on EC2, and is available in all AWS commercial and GovCloud (US) regions. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### How do I run multiple containers on the same GPU using Amazon ECS? Amazon ECS now supports GPU sharing through NVIDIA Multi-Process Service (MPS), allowing multiple ECS tasks to run concurrently on a single NVIDIA GPU. You define the GPU memory each container needs and can optionally cap the maximum share of GPU compute in the task definition, then create a service on a capacity provider configured with NVIDIA GPU instances. It works on ECS Managed Instances and ECS on EC2. _Teams packing GPU workloads more densely can track AWS infrastructure updates like this on daily.dev._ ### What happens if the NVIDIA MPS daemon crashes on an Amazon ECS Managed Instance? Amazon ECS monitors the MPS control daemon on Managed Instances, and if it stops or becomes unhealthy, ECS automatically replaces the instance and moves the affected tasks to healthy capacity. This happens without manual intervention, so GPU-sharing workloads keep running even if the daemon that enables sharing fails. _Anyone relying on GPU sharing for production workloads can follow this kind of reliability detail on daily.dev._ ## Similar posts on daily.dev - [Amazon ECS now supports fractional GPU scheduling with Amazon EC2 G6f instances](https://daily.dev/posts/amazon-ecs-now-supports-fractional-gpu-scheduling-with-amazon-ec2-g6f-instances-delytfhon) · AWS · 0 upvotes · 0 comments - [Scaling Small LLMs with NVIDIA MPS](https://daily.dev/posts/scaling-small-llms-with-nvidia-mps-0ynja1gh5) · databricks · 2 upvotes · 0 comments - [Amazon ECS Managed Instances reduces GPU management fees by up to 60%](https://daily.dev/posts/amazon-ecs-managed-instances-reduces-gpu-management-fees-by-up-to-60--pe9bchiin) · AWS · 0 upvotes · 0 comments - [GPUs on Kubernetes are bring-your-own-everything, starting with the bill](https://daily.dev/posts/gpus-on-kubernetes-are-bring-your-own-everything-starting-with-the-bill-7wb27xeuf) · The New Stack · 0 upvotes · 0 comments - [Boost GPU Memory Performance with No Code Changes Using NVIDIA CUDA MPS](https://daily.dev/posts/boost-gpu-memory-performance-with-no-code-changes-using-nvidia-cuda-mps-plqqpjajz) · NVIDIA Developer · 1 upvotes · 0 comments --- Tags: [#aws](https://daily.dev/tags/aws), [#nvidia](https://daily.dev/tags/nvidia), [#gpu](https://daily.dev/tags/gpu), [#orchestration](https://daily.dev/tags/orchestration) [View this post on daily.dev](https://daily.dev/posts/amazon-ecs-now-supports-gpu-sharing-with-nvidia-multi-process-service-mps--dbsrt4pms) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Amazon ECS now supports GPU sharing with NVIDIA Multi-Process Service (MPS)","url":"https://daily.dev/posts/amazon-ecs-now-supports-gpu-sharing-with-nvidia-multi-process-service-mps--dbsrt4pms","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/amazon-ecs-now-supports-gpu-sharing-with-nvidia-multi-process-service-mps--dbsrt4pms"},"datePublished":"2026-10-28T01:05:09.715Z","dateModified":"2026-10-28T03:56:52.582Z","description":"Amazon ECS now supports GPU sharing through NVIDIA Multi-Process Service, letting multiple ECS tasks run concurrently on a single GPU. This helps increase GPU...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/e82c8cceed8771f89f06dfc50fa9305d?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/e82c8cceed8771f89f06dfc50fa9305d?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"AWS","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"AWS","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/aws","url":"https://daily.dev/sources/aws"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/amazon-ecs-now-supports-gpu-sharing-with-nvidia-multi-process-service-mps--dbsrt4pms","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"aws,nvidia,gpu,orchestration","timeRequired":"PT2M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"AWS","item":"https://daily.dev/sources/aws"},{"@type":"ListItem","position":3,"name":"Amazon ECS now supports GPU sharing with NVIDIA Multi-Process Service (MPS)"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/amazon-ecs-now-supports-gpu-sharing-with-nvidia-multi-process-service-mps--dbsrt4pms#faq","mainEntity":[{"@type":"Question","name":"How do I run multiple containers on the same GPU using Amazon ECS?","acceptedAnswer":{"@type":"Answer","text":"Amazon ECS now supports GPU sharing through NVIDIA Multi-Process Service (MPS), allowing multiple ECS tasks to run concurrently on a single NVIDIA GPU. You define the GPU memory each container needs and can optionally cap the maximum share of GPU compute in the task definition, then create a service on a capacity provider configured with NVIDIA GPU instances. It works on ECS Managed Instances and ECS on EC2. Teams packing GPU workloads more densely can track AWS infrastructure updates like this on daily.dev."}},{"@type":"Question","name":"What happens if the NVIDIA MPS daemon crashes on an Amazon ECS Managed Instance?","acceptedAnswer":{"@type":"Answer","text":"Amazon ECS monitors the MPS control daemon on Managed Instances, and if it stops or becomes unhealthy, ECS automatically replaces the instance and moves the affected tasks to healthy capacity. This happens without manual intervention, so GPU-sharing workloads keep running even if the daemon that enables sharing fails. Anyone relying on GPU sharing for production workloads can follow this kind of reliability detail on daily.dev."}}]} ```