Overview

Amazon SageMaker is a fully managed machine learning service. With Amazon SageMaker, data scientists and developers can quickly and easily build and train machine learning models, and directly deploy them into a production-ready hosted environment.

SageMaker provides:

  • An integrated Jupyter authoring notebook instance for easy access to your data sources for exploration and analysis, so you do not have to manage servers.
  • A common machine learning algorithms that are optimized to run efficiently against extremely large data in a distributed environment.

With native support for bring-your-own-algorithms and frameworks, Amazon SageMaker offers flexible distributed training options that adjust to your specific workflows. Deploy a model into a secure and scalable environment by launching it with a single click from the Amazon SageMaker console. Training and hosting are billed by minutes of usage, with no minimum fees and no upfront commitments.

Amazon SageMaker Ground Truth

High-quality training datasets by using workers including machine learning to create labeled datasets.

See GroundTruth metrics.

Amazon SageMaker Training

An Amazon SageMaker training job is an iterative process that teaches a model to make predictions by presenting examples from a training dataset. Typically, a training algorithm computes several metrics, such as training error and prediction accuracy. These metrics help diagnose whether the model is learning well and will generalize well for making predictions on unseen data. The training algorithm writes the values of these metrics to logs, which Amazon SageMaker monitors and sends to Amazon CloudWatch in real-time.

See Training metrics.

Amazon SageMaker Endpoint

Creates an endpoint using the endpoint configuration specified in the request. Amazon SageMaker uses the endpoint to provision resources and deploy models.

Amazon SageMaker Transform Job

Use batch transform when you need to do the following:

  • Preprocess datasets to remove noise or bias that interferes with training or inference from your dataset.
  • Get inferences from large datasets.
  • Run inference when you do not need a persistent endpoint.
  • Associate input records with inferences to help interpretation results.

External reference

Amazon SageMaker

Setup

To set up the integration:

  1. Select SageMaker GroundTruth in AWS Integration Discovery Profile to discover AWS SageMaker GroundTruth.
  2. Select SageMaker Training in AWS Integration Discovery Profile to discover AWS SageMaker Training Job.
  3. Select SageMaker EndPoint in AWS Integration Discovery Profile to discover AWS SageMaker Endpoint.
  4. Select SageMaker Transform Job in AWS Integration Discovery Profile to discover AWS SageMaker Transform Job.

Event support

CloudTrail event support

  • Supported (Sagemaker GroundTruth, Training, Endpoint, Transform Job)
  • Configurable in OpsRamp AWS Integration Discovery Profile.

CloudWatch alarm support

  • Not Supported

Supported metrics

GroundTruth metrics

OpsRamp MetricAWS MetricMetric Display NameUnitAggregation TypeDescription
aws_sagemaker_labelingjobs_ActiveWorkersActiveWorkersActive WorkersNoneSumA single active worker on a private work team submitted, released, or declined a task. Use Sum to get the total number of active workers
aws_sagemaker_labelingjobs_DatasetObjectsAutoAnnotatedDatasetObjectsAutoAnnotatedDataset Objects Auto AnnotatedNoneMaximumThe number of dataset objects auto-annotated in a labeling job. Only emitted when automated labeling is enabled. Use Max to view labeling job progress
aws_sagemaker_labelingjobs_DatasetObjectsHumanAnnotatedDatasetObjectsHumanAnnotatedDataset Objects Human AnnotatedNoneMaximumThe number of dataset objects annotated by a human in a labeling job. Use Max to view labeling job progress
aws_sagemaker_labelingjobs_DatasetObjectsLabelingFailedDatasetObjectsLabelingFailedDataset Objects Labeling FailedNoneMaximumThe number of dataset objects that failed labeling in a labeling job. Use Max to view labeling job progress
aws_sagemaker_labelingjobs_JobsFailedJobsFailedJobs FailedNoneSumA single labeling job failed. Use Sum to get the total number of failed labeling jobs
aws_sagemaker_labelingjobs_JobsSucceededJobsSucceededJobs SucceededNoneSumA single labeling job succeeded. Use Sum to get the total number of succeeded labeling jobs
aws_sagemaker_labelingjobs_JobsStoppedJobsStoppedJobs StoppedNoneSumA single labeling job was stopped. Use Sum to get the total number of stopped labeling jobs
aws_sagemaker_labelingjobs_TasksAcceptedTasksAcceptedTasks AcceptedNoneSumA single task was accepted by a worker. Use Sum for total accepted tasks
aws_sagemaker_labelingjobs_TasksDeclinedTasksDeclinedTasks DeclinedNoneSumA single task was declined by a worker. Use Sum for total declined tasks
aws_sagemaker_labelingjobs_TasksReturnedTasksReturnedTasks ReturnedNoneSumA single task was returned. Use Sum for total returned tasks
aws_sagemaker_labelingjobs_TasksSubmittedTasksSubmittedTasks SubmittedNoneSumA single task was submitted/completed by a private worker. Use Sum for total submitted tasks
aws_sagemaker_labelingjobs_TimeSpentTimeSpentTime SpentSecondsSumTime spent on a task completed by a private worker. Does not include time when a worker paused or took a break
aws_sagemaker_labelingjobs_TotalDatasetObjectsLabeledTotalDatasetObjectsLabeledTotal Dataset Objects LabeledNoneMaximumThe number of dataset objects labeled successfully in a labeling job. Use Max to view labeling job progress

Training metrics

OpsRamp MetricAWS MetricMetric Display NameUnitAggregation TypeDescription
aws_sagemaker_trainingjobs_CPUUtilizationCPUUtilizationCPU UtilizationPercentAverageThe sum of each individual CPU core utilization. Each core ranges 0-100. For processing jobs: CPU utilization of the processing container
aws_sagemaker_trainingjobs_MemoryUtilizationMemoryUtilizationMemory UtilizationPercentAverageThe percentage of memory used by the containers on an instance. Range 0%-100%
aws_sagemaker_trainingjobs_GPUUtilizationGPUUtilizationGPU UtilizationPercentAverageThe percentage of GPU units used by the containers on an instance. Range 0-100 multiplied by the number of GPUs
aws_sagemaker_trainingjobs_GPUMemoryUtilizationGPUMemoryUtilizationGPU Memory UtilizationPercentAverageThe percentage of GPU memory used by the containers on an instance. Range 0-100 multiplied by number of GPUs
aws_sagemaker_trainingjobs_DiskUtilizationDiskUtilizationDisk UtilizationPercentAverageThe percentage of disk space used by the containers on an instance. Range 0%-100%