From c1061c84abbe373709b632e44b062d49b6ba3c0e Mon Sep 17 00:00:00 2001 From: Luke de Oliveira Date: Sun, 20 Nov 2016 23:19:16 -0800 Subject: [PATCH 1/2] add info about GPU-focused AMI --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index f7e235d..a3a0c8a 100644 --- a/README.md +++ b/README.md @@ -757,7 +757,7 @@ EC2 - Avoid sharing keys and [add individual ssh keys](http://security.stackexchange.com/questions/87480/managing-multiple-ssh-private-keys-for-a-team) for individual users. - **GPU support:** You can rent GPU-enabled instances on EC2 for use in machine learning or graphics rendering workloads. - There are [three generations](http://docs.aws.amazon.com/AWSEC2/latest/UserGuide/using_cluster_computing.html) of GPU-enabled instances available: - - Third generation P2 series offers NVIDIA K80 GPUs in 1, 8 and 16 GPU configurations targeting machine learning and scientific workloads. + - Third generation P2 series offers NVIDIA K80 GPUs in 1, 8 and 16 GPU configurations targeting machine learning and scientific workloads. AWS offers an [AMI](https://aws.amazon.com/marketplace/pp/B01M0AXXQB?qid=1475211685369&sr=0-1&ref_=srh_res_product_title) (based on Amazon Linux) with most NVIDIA drivers and ancillary software (CUDA, CuDNN, TensorFlow) installed. - Second generation G2 series offers NVIDIA K520 GPUs in 1 or 4 GPU configurations targeting graphics and video encoding. - First generation CG1 instances are still available in some regions in a single configuration with a NVIDIA M2050 GPU. - 🔹As with any expensive EC2 instance types, [Spot instances can offer significant savings](#ec2-cost-management) with GPU workloads when interruptions are tolerable. From 72187226694d63865fe168686555587edbe2e77d Mon Sep 17 00:00:00 2001 From: Luke de Oliveira Date: Mon, 21 Nov 2016 16:14:40 -0800 Subject: [PATCH 2/2] convert some spaces => tabs, move bullet point one level up, and add lock-in info --- README.md | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/README.md b/README.md index a3a0c8a..6ddadc4 100644 --- a/README.md +++ b/README.md @@ -756,11 +756,13 @@ EC2 - Avoid reusing the original boot keys except by administrators when creating new instances. - Avoid sharing keys and [add individual ssh keys](http://security.stackexchange.com/questions/87480/managing-multiple-ssh-private-keys-for-a-team) for individual users. - **GPU support:** You can rent GPU-enabled instances on EC2 for use in machine learning or graphics rendering workloads. - - There are [three generations](http://docs.aws.amazon.com/AWSEC2/latest/UserGuide/using_cluster_computing.html) of GPU-enabled instances available: - - Third generation P2 series offers NVIDIA K80 GPUs in 1, 8 and 16 GPU configurations targeting machine learning and scientific workloads. AWS offers an [AMI](https://aws.amazon.com/marketplace/pp/B01M0AXXQB?qid=1475211685369&sr=0-1&ref_=srh_res_product_title) (based on Amazon Linux) with most NVIDIA drivers and ancillary software (CUDA, CuDNN, TensorFlow) installed. + + - There are [three generations](http://docs.aws.amazon.com/AWSEC2/latest/UserGuide/using_cluster_computing.html) of GPU-enabled instances available: + - Third generation P2 series offers NVIDIA K80 GPUs in 1, 8 and 16 GPU configurations targeting machine learning and scientific workloads. - Second generation G2 series offers NVIDIA K520 GPUs in 1 or 4 GPU configurations targeting graphics and video encoding. - First generation CG1 instances are still available in some regions in a single configuration with a NVIDIA M2050 GPU. - - 🔹As with any expensive EC2 instance types, [Spot instances can offer significant savings](#ec2-cost-management) with GPU workloads when interruptions are tolerable. + - ⛓ AWS offers an [AMI](https://aws.amazon.com/marketplace/pp/B01M0AXXQB?qid=1475211685369&sr=0-1&ref_=srh_res_product_title) (based on Amazon Linux) with most NVIDIA drivers and ancillary software (CUDA, CUBLAS, CuDNN, TensorFlow) installed to lower the barrier to usage. Note, however, that this leads to lock-in due to Amazon Linux and the fact that you have no direct access to software configuration or versioning. + - 🔹As with any expensive EC2 instance types, [Spot instances can offer significant savings](#ec2-cost-management) with GPU workloads when interruptions are tolerable. ### EC2 Gotchas and Limitations