Class InferenceComponentSpecificationComputeResourceRequirements


  • public final class InferenceComponentSpecificationComputeResourceRequirements
    extends java.lang.Object
    • Method Detail

      • maxMemoryRequiredInMb

        public java.util.Optional<java.lang.Integer> maxMemoryRequiredInMb()
        Returns:
        Maximum MB of memory to allocate to run a model. Minimum value of `128`.
      • minMemoryRequiredInMb

        public java.lang.Integer minMemoryRequiredInMb()
        Returns:
        Minimum MB of memory to allocate to run a model. Minimum value of `128`.
      • numberOfAcceleratorDevicesRequired

        public java.util.Optional<java.lang.Double> numberOfAcceleratorDevicesRequired()
        Returns:
        Number of accelerators (GPUs or AWS Inferentia) to allocate. Minimum value of `1`.
      • numberOfCpuCoresRequired

        public java.util.Optional<java.lang.Double> numberOfCpuCoresRequired()
        Returns:
        Number of CPU cores to allocate. Minimum value of `0.25`.