FleetComputeType

Information about the compute resources the compute fleet uses. Available values include:

• ATTRIBUTE_BASED_COMPUTE : Specify the amount of vCPUs, memory, disk space, and the type of machine.

If you use ATTRIBUTE_BASED_COMPUTE , you must define your attributes by using computeConfiguration . AWS CodeBuild will select the cheapest instance that satisfies your specified attributes. For more information, see Reserved capacity environment types in the AWS CodeBuild User Guide .

• BUILD_GENERAL1_SMALL : Use up to 4 GiB memory and 2 vCPUs for builds.

• BUILD_GENERAL1_MEDIUM : Use up to 8 GiB memory and 4 vCPUs for builds.

• BUILD_GENERAL1_LARGE : Use up to 16 GiB memory and 8 vCPUs for builds, depending on your environment type.

• BUILD_GENERAL1_XLARGE : Use up to 72 GiB memory and 36 vCPUs for builds, depending on your environment type.

• BUILD_GENERAL1_2XLARGE : Use up to 144 GiB memory, 72 vCPUs, and 824 GB of SSD storage for builds. This compute type supports Docker images up to 100 GB uncompressed.

• BUILD_LAMBDA_1GB : Use up to 1 GiB memory for builds. Only available for environment type LINUX_LAMBDA_CONTAINER and ARM_LAMBDA_CONTAINER .

• BUILD_LAMBDA_2GB : Use up to 2 GiB memory for builds. Only available for environment type LINUX_LAMBDA_CONTAINER and ARM_LAMBDA_CONTAINER .

• BUILD_LAMBDA_4GB : Use up to 4 GiB memory for builds. Only available for environment type LINUX_LAMBDA_CONTAINER and ARM_LAMBDA_CONTAINER .

• BUILD_LAMBDA_8GB : Use up to 8 GiB memory for builds. Only available for environment type LINUX_LAMBDA_CONTAINER and ARM_LAMBDA_CONTAINER .

• BUILD_LAMBDA_10GB : Use up to 10 GiB memory for builds. Only available for environment type LINUX_LAMBDA_CONTAINER and ARM_LAMBDA_CONTAINER . If you use BUILD_GENERAL1_SMALL :

• For environment type LINUX_CONTAINER , you can use up to 4 GiB memory and 2 vCPUs for builds.

• For environment type LINUX_GPU_CONTAINER , you can use up to 16 GiB memory, 4 vCPUs, and 1 NVIDIA A10G Tensor Core GPU for builds.

• For environment type ARM_CONTAINER , you can use up to 4 GiB memory and 2 vCPUs on ARM-based processors for builds. If you use BUILD_GENERAL1_LARGE :

• For environment type LINUX_CONTAINER , you can use up to 16 GiB memory and 8 vCPUs for builds.

• For environment type LINUX_GPU_CONTAINER , you can use up to 255 GiB memory, 32 vCPUs, and 4 NVIDIA Tesla V100 GPUs for builds.

• For environment type ARM_CONTAINER , you can use up to 16 GiB memory and 8 vCPUs on ARM-based processors for builds. For more information, see On-demand environment types in the AWS CodeBuild User Guide.