From 36096da368c7986ebdb02df6ac7939e4ca79fe75 Mon Sep 17 00:00:00 2001 From: Kevin Pouget Date: Tue, 14 Jun 2022 17:24:27 +0200 Subject: [PATCH] DGX-A100_GPU-Operator_testing/blogpost.md: fix typo --- DGX-A100_GPU-Operator_testing/blogpost.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/DGX-A100_GPU-Operator_testing/blogpost.md b/DGX-A100_GPU-Operator_testing/blogpost.md index d67da99..b04c9a8 100644 --- a/DGX-A100_GPU-Operator_testing/blogpost.md +++ b/DGX-A100_GPU-Operator_testing/blogpost.md @@ -765,7 +765,7 @@ single AI/ML model, or all performing independent computations. As a follow-up of this work, we're planning on doing more work around AI/ML computing at large scale, such as multi-GPU multi-node training, -with multiple DGXA100 interconnected with NVIDIA GPUDirec RDMA +with multiple DGXA100 interconnected with NVIDIA GPUDirect RDMA high-performance networks. To generate enough compute requirements, we'll turn towards the greedy natural language models such as BERT or Transformer-XL. Stay tuned!