Amazon Elastic Compute Cloud Update: Added support for new accelerator types ("media") and accelerator names ("L4", "L40s", "GAUDI_HL_205", "INFERENTIA2", "TRAINIUM", "TRAINIUM2", "U30") in Attributes Based Instance Type Selection for launched instance types.

Author: AWS

.changes/next-release/feature-AmazonElasticComputeCloud-b623fb8.json

@@ -0,0 +1,6 @@
+{
+    "type": "feature",
+    "category": "Amazon Elastic Compute Cloud",
+    "contributor": "",
+    "description": "Added support for new accelerator types (\"media\") and accelerator names (\"L4\", \"L40s\", \"GAUDI_HL_205\", \"INFERENTIA2\", \"TRAINIUM\", \"TRAINIUM2\", \"U30\") in Attributes Based Instance Type Selection for launched instance types."
+}

services/ec2/src/main/resources/codegen-resources/service-2.json

@@ -7194,7 +7194,14 @@
         "v100",
         "a10g",
         "h100",
-        "t4g"
+        "t4g",
+        "l40s",
+        "l4",
+        "gaudi-hl-205",
+        "inferentia2",
+        "trainium",
+        "trainium2",
+        "u30"
       ]
     },
     "AcceleratorNameSet":{
@@ -7239,7 +7246,8 @@
       "enum":[
         "gpu",
         "fpga",
-        "inference"
+        "inference",
+        "media"
       ]
     },
     "AcceleratorTypeSet":{
@@ -42647,7 +42655,7 @@
         },
         "AcceleratorTypes":{
           "shape":"AcceleratorTypeSet",
-          "documentation":"<p>The accelerator types that must be on the instance type.</p> <ul> <li> <p>For instance types with FPGA accelerators, specify <code>fpga</code>.</p> </li> <li> <p>For instance types with GPU accelerators, specify <code>gpu</code>.</p> </li> <li> <p>For instance types with Inference accelerators, specify <code>inference</code>.</p> </li> </ul> <p>Default: Any accelerator type</p>",
+          "documentation":"<p>The accelerator types that must be on the instance type.</p> <ul> <li> <p>For instance types with FPGA accelerators, specify <code>fpga</code>.</p> </li> <li> <p>For instance types with GPU accelerators, specify <code>gpu</code>.</p> </li> <li> <p>For instance types with Inference accelerators, specify <code>inference</code>.</p> </li> <li> <p>For instance types with Media accelerators, specify <code>media</code>.</p> </li> </ul> <p>Default: Any accelerator type</p>",
           "locationName":"acceleratorTypeSet"
         },
         "AcceleratorCount":{
@@ -42662,7 +42670,7 @@
         },
         "AcceleratorNames":{
           "shape":"AcceleratorNameSet",
-          "documentation":"<p>The accelerators that must be on the instance type.</p> <ul> <li> <p>For instance types with NVIDIA A10G GPUs, specify <code>a10g</code>.</p> </li> <li> <p>For instance types with NVIDIA A100 GPUs, specify <code>a100</code>.</p> </li> <li> <p>For instance types with NVIDIA H100 GPUs, specify <code>h100</code>.</p> </li> <li> <p>For instance types with Amazon Web Services Inferentia chips, specify <code>inferentia</code>.</p> </li> <li> <p>For instance types with NVIDIA GRID K520 GPUs, specify <code>k520</code>.</p> </li> <li> <p>For instance types with NVIDIA K80 GPUs, specify <code>k80</code>.</p> </li> <li> <p>For instance types with NVIDIA M60 GPUs, specify <code>m60</code>.</p> </li> <li> <p>For instance types with AMD Radeon Pro V520 GPUs, specify <code>radeon-pro-v520</code>.</p> </li> <li> <p>For instance types with NVIDIA T4 GPUs, specify <code>t4</code>.</p> </li> <li> <p>For instance types with NVIDIA T4G GPUs, specify <code>t4g</code>.</p> </li> <li> <p>For instance types with Xilinx VU9P FPGAs, specify <code>vu9p</code>.</p> </li> <li> <p>For instance types with NVIDIA V100 GPUs, specify <code>v100</code>.</p> </li> </ul> <p>Default: Any accelerator</p>",
+          "documentation":"<p>The accelerators that must be on the instance type.</p> <ul> <li> <p>For instance types with NVIDIA A10G GPUs, specify <code>a10g</code>.</p> </li> <li> <p>For instance types with NVIDIA A100 GPUs, specify <code>a100</code>.</p> </li> <li> <p>For instance types with NVIDIA H100 GPUs, specify <code>h100</code>.</p> </li> <li> <p>For instance types with Amazon Web Services Inferentia chips, specify <code>inferentia</code>.</p> </li> <li> <p>For instance types with Amazon Web Services Inferentia2 chips, specify <code>inferentia2</code>.</p> </li> <li> <p>For instance types with Habana Gaudi HL-205 GPUs, specify <code>gaudi-hl-205</code>.</p> </li> <li> <p>For instance types with NVIDIA GRID K520 GPUs, specify <code>k520</code>.</p> </li> <li> <p>For instance types with NVIDIA K80 GPUs, specify <code>k80</code>.</p> </li> <li> <p>For instance types with NVIDIA L4 GPUs, specify <code>l4</code>.</p> </li> <li> <p>For instance types with NVIDIA L40S GPUs, specify <code>l40s</code>.</p> </li> <li> <p>For instance types with NVIDIA M60 GPUs, specify <code>m60</code>.</p> </li> <li> <p>For instance types with AMD Radeon Pro V520 GPUs, specify <code>radeon-pro-v520</code>.</p> </li> <li> <p>For instance types with Amazon Web Services Trainium chips, specify <code>trainium</code>.</p> </li> <li> <p>For instance types with Amazon Web Services Trainium2 chips, specify <code>trainium2</code>.</p> </li> <li> <p>For instance types with NVIDIA T4 GPUs, specify <code>t4</code>.</p> </li> <li> <p>For instance types with NVIDIA T4G GPUs, specify <code>t4g</code>.</p> </li> <li> <p>For instance types with Xilinx U30 cards, specify <code>u30</code>.</p> </li> <li> <p>For instance types with Xilinx VU9P FPGAs, specify <code>vu9p</code>.</p> </li> <li> <p>For instance types with NVIDIA V100 GPUs, specify <code>v100</code>.</p> </li> </ul> <p>Default: Any accelerator</p>",
           "locationName":"acceleratorNameSet"
         },
         "AcceleratorTotalMemoryMiB":{
@@ -42775,7 +42783,7 @@
         },
         "AcceleratorTypes":{
           "shape":"AcceleratorTypeSet",
-          "documentation":"<p>The accelerator types that must be on the instance type.</p> <ul> <li> <p>For instance types with FPGA accelerators, specify <code>fpga</code>.</p> </li> <li> <p>For instance types with GPU accelerators, specify <code>gpu</code>.</p> </li> <li> <p>For instance types with Inference accelerators, specify <code>inference</code>.</p> </li> </ul> <p>Default: Any accelerator type</p>",
+          "documentation":"<p>The accelerator types that must be on the instance type.</p> <ul> <li> <p>For instance types with FPGA accelerators, specify <code>fpga</code>.</p> </li> <li> <p>For instance types with GPU accelerators, specify <code>gpu</code>.</p> </li> <li> <p>For instance types with Inference accelerators, specify <code>inference</code>.</p> </li> <li> <p>For instance types with Media accelerators, specify <code>media</code>.</p> </li> </ul> <p>Default: Any accelerator type</p>",
           "locationName":"AcceleratorType"
         },
         "AcceleratorCount":{
@@ -42789,7 +42797,7 @@
         },
         "AcceleratorNames":{
           "shape":"AcceleratorNameSet",
-          "documentation":"<p>The accelerators that must be on the instance type.</p> <ul> <li> <p>For instance types with NVIDIA A10G GPUs, specify <code>a10g</code>.</p> </li> <li> <p>For instance types with NVIDIA A100 GPUs, specify <code>a100</code>.</p> </li> <li> <p>For instance types with NVIDIA H100 GPUs, specify <code>h100</code>.</p> </li> <li> <p>For instance types with Amazon Web Services Inferentia chips, specify <code>inferentia</code>.</p> </li> <li> <p>For instance types with NVIDIA GRID K520 GPUs, specify <code>k520</code>.</p> </li> <li> <p>For instance types with NVIDIA K80 GPUs, specify <code>k80</code>.</p> </li> <li> <p>For instance types with NVIDIA M60 GPUs, specify <code>m60</code>.</p> </li> <li> <p>For instance types with AMD Radeon Pro V520 GPUs, specify <code>radeon-pro-v520</code>.</p> </li> <li> <p>For instance types with NVIDIA T4 GPUs, specify <code>t4</code>.</p> </li> <li> <p>For instance types with NVIDIA T4G GPUs, specify <code>t4g</code>.</p> </li> <li> <p>For instance types with Xilinx VU9P FPGAs, specify <code>vu9p</code>.</p> </li> <li> <p>For instance types with NVIDIA V100 GPUs, specify <code>v100</code>.</p> </li> </ul> <p>Default: Any accelerator</p>",
+          "documentation":"<p>The accelerators that must be on the instance type.</p> <ul> <li> <p>For instance types with NVIDIA A10G GPUs, specify <code>a10g</code>.</p> </li> <li> <p>For instance types with NVIDIA A100 GPUs, specify <code>a100</code>.</p> </li> <li> <p>For instance types with NVIDIA H100 GPUs, specify <code>h100</code>.</p> </li> <li> <p>For instance types with Amazon Web Services Inferentia chips, specify <code>inferentia</code>.</p> </li> <li> <p>For instance types with Amazon Web Services Inferentia2 chips, specify <code>inferentia2</code>.</p> </li> <li> <p>For instance types with Habana Gaudi HL-205 GPUs, specify <code>gaudi-hl-205</code>.</p> </li> <li> <p>For instance types with NVIDIA GRID K520 GPUs, specify <code>k520</code>.</p> </li> <li> <p>For instance types with NVIDIA K80 GPUs, specify <code>k80</code>.</p> </li> <li> <p>For instance types with NVIDIA L4 GPUs, specify <code>l4</code>.</p> </li> <li> <p>For instance types with NVIDIA L40S GPUs, specify <code>l40s</code>.</p> </li> <li> <p>For instance types with NVIDIA M60 GPUs, specify <code>m60</code>.</p> </li> <li> <p>For instance types with AMD Radeon Pro V520 GPUs, specify <code>radeon-pro-v520</code>.</p> </li> <li> <p>For instance types with Amazon Web Services Trainium chips, specify <code>trainium</code>.</p> </li> <li> <p>For instance types with Amazon Web Services Trainium2 chips, specify <code>trainium2</code>.</p> </li> <li> <p>For instance types with NVIDIA T4 GPUs, specify <code>t4</code>.</p> </li> <li> <p>For instance types with NVIDIA T4G GPUs, specify <code>t4g</code>.</p> </li> <li> <p>For instance types with Xilinx U30 cards, specify <code>u30</code>.</p> </li> <li> <p>For instance types with Xilinx VU9P FPGAs, specify <code>vu9p</code>.</p> </li> <li> <p>For instance types with NVIDIA V100 GPUs, specify <code>v100</code>.</p> </li> </ul> <p>Default: Any accelerator</p>",
           "locationName":"AcceleratorName"
         },
         "AcceleratorTotalMemoryMiB":{