Developer documentation > API reference > REST API > Monitoring > Service health

Service health¶

Use the endpoints described below to manage service health. Service health tracks metrics about a deployment's ability to respond to prediction requests quickly and reliably. This helps identify bottlenecks and assess capacity, which is critical to proper provisioning.

Retrieve service health metrics by deployment ID¶

Operation path: GET /api/v2/deployments/{deploymentId}/batchServiceStats/

Authentication requirements: BearerAuth

Retrieve all deployment service health metrics over a set of batches.

Parameters

Name	In	Type	Required	Description
executionTimeQuantile	query	number	false	Quantile for executionTime metric.
responseTimeQuantile	query	number	false	Quantile for responseTime metric.
slowRequestsThreshold	query	integer	false	Threshold for slowRequests metric.
segmentAttribute	query	string	false	The name of a segment attribute used for segment analysis.
segmentValue	query	string,null	false	The value of the `segmentAttribute` to segment on.
batchId	query	any	false	The id of the batch for which metrics are being retrieved.
modelId	query	string	false	The id of the model for which metrics are being retrieved.
deploymentId	path	string	true	Unique identifier of the deployment.

Enumerated Values¶

Parameter	Value
segmentAttribute	[`DataRobot-Consumer`, `DataRobot-Remote-IP`, `DataRobot-Host-IP`]

Example responses¶

200 Response

{
  "properties": {
    "batches": {
      "description": "Info of the batches the metric is aggregated on.",
      "items": {
        "description": "Batch info.",
        "properties": {
          "earliestPredictionTimestamp": {
            "description": "Earliest prediction timestamp of a batch.",
            "format": "date-time",
            "type": "string",
            "x-versionadded": "v2.33"
          },
          "id": {
            "description": "Batch ID.",
            "type": "string",
            "x-versionadded": "v2.33"
          },
          "latestPredictionTimestamp": {
            "description": "Latest prediction timestamp of a batch.",
            "format": "date-time",
            "type": [
              "string",
              "null"
            ],
            "x-versionadded": "v2.33"
          },
          "name": {
            "description": "Batch name.",
            "type": "string",
            "x-versionadded": "v2.33"
          }
        },
        "required": [
          "earliestPredictionTimestamp",
          "id",
          "latestPredictionTimestamp",
          "name"
        ],
        "type": "object"
      },
      "type": "array",
      "x-versionadded": "v2.33"
    },
    "metrics": {
      "description": "Service health metrics of the deployment",
      "properties": {
        "cacheHitRatio": {
          "description": "Number of cache hits.",
          "type": [
            "number",
            "null"
          ],
          "x-versionadded": "v2.33"
        },
        "executionTime": {
          "description": "Request execution time at executionTimeQuantile (in milliseconds).",
          "type": [
            "number",
            "null"
          ],
          "x-versionadded": "v2.33"
        },
        "numConsumers": {
          "description": "Number of unique users performing requests.",
          "type": [
            "integer",
            "null"
          ],
          "x-versionadded": "v2.33"
        },
        "responseTime": {
          "description": "Request response time at responseTimeQuantile (in milliseconds).",
          "type": [
            "number",
            "null"
          ],
          "x-versionadded": "v2.33"
        },
        "serverErrorRate": {
          "description": "Ratio of server errors to the total number of requests.",
          "type": [
            "number",
            "null"
          ],
          "x-versionadded": "v2.33"
        },
        "slowRequests": {
          "description": "Number of requests with response time greater than slowRequestsThreshold",
          "type": [
            "integer",
            "null"
          ],
          "x-versionadded": "v2.33"
        },
        "totalPredictions": {
          "description": "Total number of prediction rows.",
          "type": [
            "integer",
            "null"
          ],
          "x-versionadded": "v2.33"
        },
        "totalRequests": {
          "description": "Total number of prediction requests performed.",
          "type": [
            "integer",
            "null"
          ],
          "x-versionadded": "v2.33"
        },
        "userErrorRate": {
          "description": "Ratio of user errors to the total number of requests.",
          "type": [
            "number",
            "null"
          ],
          "x-versionadded": "v2.33"
        }
      },
      "required": [
        "cacheHitRatio",
        "executionTime",
        "numConsumers",
        "responseTime",
        "serverErrorRate",
        "slowRequests",
        "totalPredictions",
        "totalRequests",
        "userErrorRate"
      ],
      "type": "object"
    },
    "segmentAttribute": {
      "description": "The name of the segment on which segment analysis is being performed.",
      "type": [
        "string",
        "null"
      ],
      "x-versionadded": "v2.20"
    },
    "segmentValue": {
      "default": "",
      "description": "The value of the `segmentAttribute` to segment on.",
      "type": [
        "string",
        "null"
      ],
      "x-versionadded": "v2.20"
    }
  },
  "required": [
    "batches",
    "metrics"
  ],
  "type": "object"
}

Responses¶

Status	Meaning	Description	Schema
200	OK	Service health metric data retrieved.	ServiceStatsForBatchRetrieveResponse

Retrieve service stats by id¶

Operation path: GET /api/v2/deployments/{deploymentId}/serviceStats/

Authentication requirements: BearerAuth

Retrieve all deployment service health metrics over a single period of time.

Parameters

Name	In	Type	Required	Description
start	query	string,null(date-time)	false	Start of the period to retrieve monitoring data, defaults to 7 days ago from the end of the period. Note: this field only accepts top of the hour RFC3339 datetime strings, for example: `2019-08-01T00:00:00Z`.
end	query	string,null(date-time)	false	End of the period to retrieve monitoring data, defaults to the next top of the hour. Note: this field only accepts top of the hour RFC3339 datetime strings, for example: `2019-08-01T00:00:00Z`.
executionTimeQuantile	query	number	false	Quantile for executionTime metric.
responseTimeQuantile	query	number	false	Quantile for responseTime metric.
slowRequestsThreshold	query	integer	false	Threshold for slowRequests metric.
segmentAttribute	query	string	false	The name of a segment attribute used for segment analysis.
segmentValue	query	string,null	false	The value of the `segmentAttribute` to segment on.
modelId	query	string	false	The id of the model for which metrics are being retrieved.
deploymentId	path	string	true	Unique identifier of the deployment.

Enumerated Values¶

Parameter	Value
segmentAttribute	[`DataRobot-Consumer`, `DataRobot-Remote-IP`, `DataRobot-Host-IP`]

Example responses¶

200 Response

{
  "properties": {
    "metrics": {
      "description": "Service health metrics of the deployment",
      "properties": {
        "cacheHitRatio": {
          "description": "Number of cache hits.",
          "type": [
            "number",
            "null"
          ]
        },
        "executionTime": {
          "description": "Request execution time at executionTimeQuantile (in milliseconds).",
          "type": [
            "number",
            "null"
          ]
        },
        "medianLoad": {
          "description": "Median of the request rate (in requests per minute).",
          "type": [
            "number",
            "null"
          ]
        },
        "numConsumers": {
          "description": "Number of unique users performing requests.",
          "type": [
            "integer",
            "null"
          ]
        },
        "peakLoad": {
          "description": "Maximum of the request rate (in requests per minute).",
          "type": [
            "number",
            "null"
          ]
        },
        "responseTime": {
          "description": "Request response time at responseTimeQuantile (in milliseconds).",
          "type": [
            "number",
            "null"
          ]
        },
        "serverErrorRate": {
          "description": "Ratio of server errors to the total number of requests.",
          "type": [
            "number",
            "null"
          ]
        },
        "slowRequests": {
          "description": "Number of requests with response time greater than slowRequestsThreshold",
          "type": [
            "integer",
            "null"
          ]
        },
        "totalPredictions": {
          "description": "Total number of prediction rows.",
          "type": [
            "integer",
            "null"
          ]
        },
        "totalRequests": {
          "description": "Total number of prediction requests performed.",
          "type": [
            "integer",
            "null"
          ]
        },
        "userErrorRate": {
          "description": "Ratio of user errors to the total number of requests.",
          "type": [
            "number",
            "null"
          ]
        }
      },
      "required": [
        "cacheHitRatio",
        "executionTime",
        "medianLoad",
        "numConsumers",
        "peakLoad",
        "responseTime",
        "serverErrorRate",
        "slowRequests",
        "totalPredictions",
        "totalRequests",
        "userErrorRate"
      ],
      "type": "object"
    },
    "modelId": {
      "description": "The id of the model for which metrics are being retrieved.",
      "type": "string"
    },
    "period": {
      "description": "An object with the keys \"start\" and \"end\" defining the period.",
      "properties": {
        "end": {
          "description": "End of the period to retrieve monitoring data, defaults to the next top of the hour. Note: this field only accepts top of the hour RFC3339 datetime strings, for example: ``2019-08-01T00:00:00Z``.",
          "format": "date-time",
          "type": [
            "string",
            "null"
          ]
        },
        "start": {
          "description": "Start of the period to retrieve monitoring data, defaults to 7 days ago from the end of the period. Note: this field only accepts top of the hour RFC3339 datetime strings, for example: ``2019-08-01T00:00:00Z``.",
          "format": "date-time",
          "type": [
            "string",
            "null"
          ]
        }
      },
      "type": "object"
    },
    "segmentAttribute": {
      "description": "The name of the segment on which segment analysis is being performed.",
      "type": [
        "string",
        "null"
      ],
      "x-versionadded": "v2.20"
    },
    "segmentValue": {
      "default": "",
      "description": "The value of the `segmentAttribute` to segment on.",
      "type": [
        "string",
        "null"
      ],
      "x-versionadded": "v2.20"
    }
  },
  "required": [
    "metrics",
    "period"
  ],
  "type": "object"
}

Responses¶

Status	Meaning	Description	Schema
200	OK	Service health metric data retrieved.	ServiceStatsForTimeRangeResponse

Retrieve service health metric over batch by deployment ID¶

Operation path: GET /api/v2/deployments/{deploymentId}/serviceStatsOverBatch/

Authentication requirements: BearerAuth

Retrieve values for one single deployment service health metric over batch.

Parameters

Name	In	Type	Required	Description
segmentAttribute	query	string	false	The name of a segment attribute used for segment analysis.
segmentValue	query	string,null	false	The value of the `segmentAttribute` to segment on.
batchId	query	any	false	The id of the batch for which metrics are being retrieved.
modelId	query	string	false	The id of the model for which metrics are being retrieved.
metric	query	string	false	A service health metric.
quantile	query	number	false	Quantile for executionTime and responseTime metrics
threshold	query	integer	false	Threshold for slowQueries metric.
deploymentId	path	string	true	Unique identifier of the deployment.

Enumerated Values¶

Parameter	Value
segmentAttribute	[`DataRobot-Consumer`, `DataRobot-Remote-IP`, `DataRobot-Host-IP`]
metric	[`totalPredictions`, `totalRequests`, `slowRequests`, `executionTime`, `responseTime`, `userErrorRate`, `serverErrorRate`, `numConsumers`, `cacheHitRatio`]

Example responses¶

200 Response

{
  "properties": {
    "buckets": {
      "description": "An array of buckets, representing service health stats of the deployment over selected batches.",
      "items": {
        "description": "Service health stats of the deployment over a batch.",
        "properties": {
          "batch": {
            "description": "Batch info.",
            "properties": {
              "earliestPredictionTimestamp": {
                "description": "Earliest prediction timestamp of a batch.",
                "format": "date-time",
                "type": "string",
                "x-versionadded": "v2.33"
              },
              "id": {
                "description": "Batch ID.",
                "type": "string",
                "x-versionadded": "v2.33"
              },
              "latestPredictionTimestamp": {
                "description": "Latest prediction timestamp of a batch.",
                "format": "date-time",
                "type": [
                  "string",
                  "null"
                ],
                "x-versionadded": "v2.33"
              },
              "name": {
                "description": "Batch name.",
                "type": "string",
                "x-versionadded": "v2.33"
              }
            },
            "required": [
              "earliestPredictionTimestamp",
              "id",
              "latestPredictionTimestamp",
              "name"
            ],
            "type": "object"
          },
          "value": {
            "description": "Value of the metric in the bucket.",
            "type": [
              "number",
              "null"
            ],
            "x-versionadded": "v2.33"
          }
        },
        "required": [
          "batch",
          "value"
        ],
        "type": "object"
      },
      "type": "array",
      "x-versionadded": "v2.33"
    },
    "metric": {
      "description": "Name of the metric requested.",
      "type": "string",
      "x-versionadded": "v2.33"
    },
    "segmentAttribute": {
      "description": "The name of the segment on which segment analysis is being performed.",
      "type": [
        "string",
        "null"
      ],
      "x-versionadded": "v2.20"
    },
    "segmentValue": {
      "default": "",
      "description": "The value of the `segmentAttribute` to segment on.",
      "type": [
        "string",
        "null"
      ],
      "x-versionadded": "v2.20"
    }
  },
  "required": [
    "buckets",
    "metric"
  ],
  "type": "object"
}

Responses¶

Status	Meaning	Description	Schema
200	OK	Service health metric data retrieved.	ServiceStatsOverBatchResponse

Retrieve service health metric over time by deployment ID¶

Operation path: GET /api/v2/deployments/{deploymentId}/serviceStatsOverTime/

Authentication requirements: BearerAuth

Retrieve values for one single deployment service health metric over time.

Parameters

Name	In	Type	Required	Description
start	query	string,null(date-time)	false	Start of the period to retrieve monitoring data, defaults to 7 days ago from the end of the period. Note: this field only accepts top of the hour RFC3339 datetime strings, for example: `2019-08-01T00:00:00Z`.
end	query	string,null(date-time)	false	End of the period to retrieve monitoring data, defaults to the next top of the hour. Note: this field only accepts top of the hour RFC3339 datetime strings, for example: `2019-08-01T00:00:00Z`.
bucketSize	query	string(duration)	false	The time duration of a bucket. Needs to be multiple of one hour. Can not be longer than the total length of the period. If not set, a default value will be calculated based on the start and end time.
segmentAttribute	query	string	false	The name of a segment attribute used for segment analysis.
segmentValue	query	string,null	false	The value of the `segmentAttribute` to segment on.
modelId	query	any	false	The ID of the models for which metrics are being retrieved.
metric	query	string	false	Name of the metric. See below for a list of supported metrics.
quantile	query	number	false	A quantile for resulting data, used if `metric` is `executionTime` or `responseTime`, defaults to 0.5.
threshold	query	integer	false	A threshold for filtering results, used if `metric` is `slowQueries`, defaults to 1000.
deploymentId	path	string	true	Unique identifier of the deployment.

Enumerated Values¶

Parameter	Value
segmentAttribute	[`DataRobot-Consumer`, `DataRobot-Remote-IP`, `DataRobot-Host-IP`]
metric	[`totalPredictions`, `totalRequests`, `slowRequests`, `executionTime`, `responseTime`, `userErrorRate`, `serverErrorRate`, `numConsumers`, `cacheHitRatio`, `medianLoad`, `peakLoad`]

Example responses¶

200 Response

{
  "properties": {
    "buckets": {
      "description": "An array of bucket, representing service health stats of the deployment over time.",
      "items": {
        "description": "Service health stats of the deployment over a time range.",
        "properties": {
          "modelId": {
            "description": "The id of the model for which metrics are being retrieved.",
            "type": "string",
            "x-versionadded": "v2.37"
          },
          "period": {
            "description": "An object with the keys \"start\" and \"end\" defining the period.",
            "properties": {
              "end": {
                "description": "End of the period to retrieve monitoring data, defaults to the next top of the hour. Note: this field only accepts top of the hour RFC3339 datetime strings, for example: ``2019-08-01T00:00:00Z``.",
                "format": "date-time",
                "type": [
                  "string",
                  "null"
                ]
              },
              "start": {
                "description": "Start of the period to retrieve monitoring data, defaults to 7 days ago from the end of the period. Note: this field only accepts top of the hour RFC3339 datetime strings, for example: ``2019-08-01T00:00:00Z``.",
                "format": "date-time",
                "type": [
                  "string",
                  "null"
                ]
              }
            },
            "type": "object"
          },
          "value": {
            "description": "Value of the metric in the bucket. Null if no value",
            "type": [
              "number",
              "null"
            ]
          }
        },
        "required": [
          "period",
          "value"
        ],
        "type": "object"
      },
      "type": "array"
    },
    "metric": {
      "description": "Name of the metric requested.",
      "type": "string"
    },
    "modelId": {
      "description": "The id of the model for which metrics are being retrieved. Deprecated, use modelId in each bucket instead.",
      "type": "string",
      "x-versiondeprecated": "v2.37"
    },
    "segmentAttribute": {
      "description": "The name of the segment on which segment analysis is being performed.",
      "type": [
        "string",
        "null"
      ],
      "x-versionadded": "v2.20"
    },
    "segmentValue": {
      "default": "",
      "description": "The value of the `segmentAttribute` to segment on.",
      "type": [
        "string",
        "null"
      ],
      "x-versionadded": "v2.20"
    },
    "summary": {
      "description": "Service health stats of the deployment over a time range.",
      "properties": {
        "modelId": {
          "description": "The id of the model for which metrics are being retrieved.",
          "type": "string",
          "x-versionadded": "v2.37"
        },
        "period": {
          "description": "An object with the keys \"start\" and \"end\" defining the period.",
          "properties": {
            "end": {
              "description": "End of the period to retrieve monitoring data, defaults to the next top of the hour. Note: this field only accepts top of the hour RFC3339 datetime strings, for example: ``2019-08-01T00:00:00Z``.",
              "format": "date-time",
              "type": [
                "string",
                "null"
              ]
            },
            "start": {
              "description": "Start of the period to retrieve monitoring data, defaults to 7 days ago from the end of the period. Note: this field only accepts top of the hour RFC3339 datetime strings, for example: ``2019-08-01T00:00:00Z``.",
              "format": "date-time",
              "type": [
                "string",
                "null"
              ]
            }
          },
          "type": "object"
        },
        "value": {
          "description": "Value of the metric in the bucket. Null if no value",
          "type": [
            "number",
            "null"
          ]
        }
      },
      "required": [
        "period",
        "value"
      ],
      "type": "object"
    }
  },
  "required": [
    "buckets",
    "metric"
  ],
  "type": "object"
}

Responses¶

Status	Meaning	Description	Schema
200	OK	Service health metric data retrieved.	ServiceStatsOverTimeResponse

Schemas¶

Batch

{
  "description": "Batch info.",
  "properties": {
    "earliestPredictionTimestamp": {
      "description": "Earliest prediction timestamp of a batch.",
      "format": "date-time",
      "type": "string",
      "x-versionadded": "v2.33"
    },
    "id": {
      "description": "Batch ID.",
      "type": "string",
      "x-versionadded": "v2.33"
    },
    "latestPredictionTimestamp": {
      "description": "Latest prediction timestamp of a batch.",
      "format": "date-time",
      "type": [
        "string",
        "null"
      ],
      "x-versionadded": "v2.33"
    },
    "name": {
      "description": "Batch name.",
      "type": "string",
      "x-versionadded": "v2.33"
    }
  },
  "required": [
    "earliestPredictionTimestamp",
    "id",
    "latestPredictionTimestamp",
    "name"
  ],
  "type": "object"
}

Batch info.

Properties¶

Name	Type	Required	Description
earliestPredictionTimestamp	string(date-time)	true	Earliest prediction timestamp of a batch.
id	string	true	Batch ID.
latestPredictionTimestamp	string,null(date-time)	true	Latest prediction timestamp of a batch.
name	string	true	Batch name.

ServiceStatsForBatchRetrieveResponse

{
  "properties": {
    "batches": {
      "description": "Info of the batches the metric is aggregated on.",
      "items": {
        "description": "Batch info.",
        "properties": {
          "earliestPredictionTimestamp": {
            "description": "Earliest prediction timestamp of a batch.",
            "format": "date-time",
            "type": "string",
            "x-versionadded": "v2.33"
          },
          "id": {
            "description": "Batch ID.",
            "type": "string",
            "x-versionadded": "v2.33"
          },
          "latestPredictionTimestamp": {
            "description": "Latest prediction timestamp of a batch.",
            "format": "date-time",
            "type": [
              "string",
              "null"
            ],
            "x-versionadded": "v2.33"
          },
          "name": {
            "description": "Batch name.",
            "type": "string",
            "x-versionadded": "v2.33"
          }
        },
        "required": [
          "earliestPredictionTimestamp",
          "id",
          "latestPredictionTimestamp",
          "name"
        ],
        "type": "object"
      },
      "type": "array",
      "x-versionadded": "v2.33"
    },
    "metrics": {
      "description": "Service health metrics of the deployment",
      "properties": {
        "cacheHitRatio": {
          "description": "Number of cache hits.",
          "type": [
            "number",
            "null"
          ],
          "x-versionadded": "v2.33"
        },
        "executionTime": {
          "description": "Request execution time at executionTimeQuantile (in milliseconds).",
          "type": [
            "number",
            "null"
          ],
          "x-versionadded": "v2.33"
        },
        "numConsumers": {
          "description": "Number of unique users performing requests.",
          "type": [
            "integer",
            "null"
          ],
          "x-versionadded": "v2.33"
        },
        "responseTime": {
          "description": "Request response time at responseTimeQuantile (in milliseconds).",
          "type": [
            "number",
            "null"
          ],
          "x-versionadded": "v2.33"
        },
        "serverErrorRate": {
          "description": "Ratio of server errors to the total number of requests.",
          "type": [
            "number",
            "null"
          ],
          "x-versionadded": "v2.33"
        },
        "slowRequests": {
          "description": "Number of requests with response time greater than slowRequestsThreshold",
          "type": [
            "integer",
            "null"
          ],
          "x-versionadded": "v2.33"
        },
        "totalPredictions": {
          "description": "Total number of prediction rows.",
          "type": [
            "integer",
            "null"
          ],
          "x-versionadded": "v2.33"
        },
        "totalRequests": {
          "description": "Total number of prediction requests performed.",
          "type": [
            "integer",
            "null"
          ],
          "x-versionadded": "v2.33"
        },
        "userErrorRate": {
          "description": "Ratio of user errors to the total number of requests.",
          "type": [
            "number",
            "null"
          ],
          "x-versionadded": "v2.33"
        }
      },
      "required": [
        "cacheHitRatio",
        "executionTime",
        "numConsumers",
        "responseTime",
        "serverErrorRate",
        "slowRequests",
        "totalPredictions",
        "totalRequests",
        "userErrorRate"
      ],
      "type": "object"
    },
    "segmentAttribute": {
      "description": "The name of the segment on which segment analysis is being performed.",
      "type": [
        "string",
        "null"
      ],
      "x-versionadded": "v2.20"
    },
    "segmentValue": {
      "default": "",
      "description": "The value of the `segmentAttribute` to segment on.",
      "type": [
        "string",
        "null"
      ],
      "x-versionadded": "v2.20"
    }
  },
  "required": [
    "batches",
    "metrics"
  ],
  "type": "object"
}

Properties¶

Name	Type	Required	Description
batches	[Batch]	true	Info of the batches the metric is aggregated on.
metrics	ServiceStatsMetricsForBatch	true	Service health metrics of the deployment
segmentAttribute	string,null	false	The name of the segment on which segment analysis is being performed.
segmentValue	string,null	false	The value of the `segmentAttribute` to segment on.

ServiceStatsForTimeRangeResponse

{
  "properties": {
    "metrics": {
      "description": "Service health metrics of the deployment",
      "properties": {
        "cacheHitRatio": {
          "description": "Number of cache hits.",
          "type": [
            "number",
            "null"
          ]
        },
        "executionTime": {
          "description": "Request execution time at executionTimeQuantile (in milliseconds).",
          "type": [
            "number",
            "null"
          ]
        },
        "medianLoad": {
          "description": "Median of the request rate (in requests per minute).",
          "type": [
            "number",
            "null"
          ]
        },
        "numConsumers": {
          "description": "Number of unique users performing requests.",
          "type": [
            "integer",
            "null"
          ]
        },
        "peakLoad": {
          "description": "Maximum of the request rate (in requests per minute).",
          "type": [
            "number",
            "null"
          ]
        },
        "responseTime": {
          "description": "Request response time at responseTimeQuantile (in milliseconds).",
          "type": [
            "number",
            "null"
          ]
        },
        "serverErrorRate": {
          "description": "Ratio of server errors to the total number of requests.",
          "type": [
            "number",
            "null"
          ]
        },
        "slowRequests": {
          "description": "Number of requests with response time greater than slowRequestsThreshold",
          "type": [
            "integer",
            "null"
          ]
        },
        "totalPredictions": {
          "description": "Total number of prediction rows.",
          "type": [
            "integer",
            "null"
          ]
        },
        "totalRequests": {
          "description": "Total number of prediction requests performed.",
          "type": [
            "integer",
            "null"
          ]
        },
        "userErrorRate": {
          "description": "Ratio of user errors to the total number of requests.",
          "type": [
            "number",
            "null"
          ]
        }
      },
      "required": [
        "cacheHitRatio",
        "executionTime",
        "medianLoad",
        "numConsumers",
        "peakLoad",
        "responseTime",
        "serverErrorRate",
        "slowRequests",
        "totalPredictions",
        "totalRequests",
        "userErrorRate"
      ],
      "type": "object"
    },
    "modelId": {
      "description": "The id of the model for which metrics are being retrieved.",
      "type": "string"
    },
    "period": {
      "description": "An object with the keys \"start\" and \"end\" defining the period.",
      "properties": {
        "end": {
          "description": "End of the period to retrieve monitoring data, defaults to the next top of the hour. Note: this field only accepts top of the hour RFC3339 datetime strings, for example: ``2019-08-01T00:00:00Z``.",
          "format": "date-time",
          "type": [
            "string",
            "null"
          ]
        },
        "start": {
          "description": "Start of the period to retrieve monitoring data, defaults to 7 days ago from the end of the period. Note: this field only accepts top of the hour RFC3339 datetime strings, for example: ``2019-08-01T00:00:00Z``.",
          "format": "date-time",
          "type": [
            "string",
            "null"
          ]
        }
      },
      "type": "object"
    },
    "segmentAttribute": {
      "description": "The name of the segment on which segment analysis is being performed.",
      "type": [
        "string",
        "null"
      ],
      "x-versionadded": "v2.20"
    },
    "segmentValue": {
      "default": "",
      "description": "The value of the `segmentAttribute` to segment on.",
      "type": [
        "string",
        "null"
      ],
      "x-versionadded": "v2.20"
    }
  },
  "required": [
    "metrics",
    "period"
  ],
  "type": "object"
}

Properties¶

Name	Type	Required	Description
metrics	ServiceStatsMetricsForTimeRange	true	Service health metrics of the deployment
modelId	string	false	The id of the model for which metrics are being retrieved.
period	TimeRange	true	An object with the keys "start" and "end" defining the period.
segmentAttribute	string,null	false	The name of the segment on which segment analysis is being performed.
segmentValue	string,null	false	The value of the `segmentAttribute` to segment on.

ServiceStatsMetricsForBatch

{
  "description": "Service health metrics of the deployment",
  "properties": {
    "cacheHitRatio": {
      "description": "Number of cache hits.",
      "type": [
        "number",
        "null"
      ],
      "x-versionadded": "v2.33"
    },
    "executionTime": {
      "description": "Request execution time at executionTimeQuantile (in milliseconds).",
      "type": [
        "number",
        "null"
      ],
      "x-versionadded": "v2.33"
    },
    "numConsumers": {
      "description": "Number of unique users performing requests.",
      "type": [
        "integer",
        "null"
      ],
      "x-versionadded": "v2.33"
    },
    "responseTime": {
      "description": "Request response time at responseTimeQuantile (in milliseconds).",
      "type": [
        "number",
        "null"
      ],
      "x-versionadded": "v2.33"
    },
    "serverErrorRate": {
      "description": "Ratio of server errors to the total number of requests.",
      "type": [
        "number",
        "null"
      ],
      "x-versionadded": "v2.33"
    },
    "slowRequests": {
      "description": "Number of requests with response time greater than slowRequestsThreshold",
      "type": [
        "integer",
        "null"
      ],
      "x-versionadded": "v2.33"
    },
    "totalPredictions": {
      "description": "Total number of prediction rows.",
      "type": [
        "integer",
        "null"
      ],
      "x-versionadded": "v2.33"
    },
    "totalRequests": {
      "description": "Total number of prediction requests performed.",
      "type": [
        "integer",
        "null"
      ],
      "x-versionadded": "v2.33"
    },
    "userErrorRate": {
      "description": "Ratio of user errors to the total number of requests.",
      "type": [
        "number",
        "null"
      ],
      "x-versionadded": "v2.33"
    }
  },
  "required": [
    "cacheHitRatio",
    "executionTime",
    "numConsumers",
    "responseTime",
    "serverErrorRate",
    "slowRequests",
    "totalPredictions",
    "totalRequests",
    "userErrorRate"
  ],
  "type": "object"
}

Service health metrics of the deployment

Properties¶

Name	Type	Required	Description
cacheHitRatio	number,null	true	Number of cache hits.
executionTime	number,null	true	Request execution time at executionTimeQuantile (in milliseconds).
numConsumers	integer,null	true	Number of unique users performing requests.
responseTime	number,null	true	Request response time at responseTimeQuantile (in milliseconds).
serverErrorRate	number,null	true	Ratio of server errors to the total number of requests.
slowRequests	integer,null	true	Number of requests with response time greater than slowRequestsThreshold
totalPredictions	integer,null	true	Total number of prediction rows.
totalRequests	integer,null	true	Total number of prediction requests performed.
userErrorRate	number,null	true	Ratio of user errors to the total number of requests.

ServiceStatsMetricsForTimeRange

{
  "description": "Service health metrics of the deployment",
  "properties": {
    "cacheHitRatio": {
      "description": "Number of cache hits.",
      "type": [
        "number",
        "null"
      ]
    },
    "executionTime": {
      "description": "Request execution time at executionTimeQuantile (in milliseconds).",
      "type": [
        "number",
        "null"
      ]
    },
    "medianLoad": {
      "description": "Median of the request rate (in requests per minute).",
      "type": [
        "number",
        "null"
      ]
    },
    "numConsumers": {
      "description": "Number of unique users performing requests.",
      "type": [
        "integer",
        "null"
      ]
    },
    "peakLoad": {
      "description": "Maximum of the request rate (in requests per minute).",
      "type": [
        "number",
        "null"
      ]
    },
    "responseTime": {
      "description": "Request response time at responseTimeQuantile (in milliseconds).",
      "type": [
        "number",
        "null"
      ]
    },
    "serverErrorRate": {
      "description": "Ratio of server errors to the total number of requests.",
      "type": [
        "number",
        "null"
      ]
    },
    "slowRequests": {
      "description": "Number of requests with response time greater than slowRequestsThreshold",
      "type": [
        "integer",
        "null"
      ]
    },
    "totalPredictions": {
      "description": "Total number of prediction rows.",
      "type": [
        "integer",
        "null"
      ]
    },
    "totalRequests": {
      "description": "Total number of prediction requests performed.",
      "type": [
        "integer",
        "null"
      ]
    },
    "userErrorRate": {
      "description": "Ratio of user errors to the total number of requests.",
      "type": [
        "number",
        "null"
      ]
    }
  },
  "required": [
    "cacheHitRatio",
    "executionTime",
    "medianLoad",
    "numConsumers",
    "peakLoad",
    "responseTime",
    "serverErrorRate",
    "slowRequests",
    "totalPredictions",
    "totalRequests",
    "userErrorRate"
  ],
  "type": "object"
}

Service health metrics of the deployment

Properties¶

Name	Type	Required	Description
cacheHitRatio	number,null	true	Number of cache hits.
executionTime	number,null	true	Request execution time at executionTimeQuantile (in milliseconds).
medianLoad	number,null	true	Median of the request rate (in requests per minute).
numConsumers	integer,null	true	Number of unique users performing requests.
peakLoad	number,null	true	Maximum of the request rate (in requests per minute).
responseTime	number,null	true	Request response time at responseTimeQuantile (in milliseconds).
serverErrorRate	number,null	true	Ratio of server errors to the total number of requests.
slowRequests	integer,null	true	Number of requests with response time greater than slowRequestsThreshold
totalPredictions	integer,null	true	Total number of prediction rows.
totalRequests	integer,null	true	Total number of prediction requests performed.
userErrorRate	number,null	true	Ratio of user errors to the total number of requests.

ServiceStatsOverBatchBucket

{
  "description": "Service health stats of the deployment over a batch.",
  "properties": {
    "batch": {
      "description": "Batch info.",
      "properties": {
        "earliestPredictionTimestamp": {
          "description": "Earliest prediction timestamp of a batch.",
          "format": "date-time",
          "type": "string",
          "x-versionadded": "v2.33"
        },
        "id": {
          "description": "Batch ID.",
          "type": "string",
          "x-versionadded": "v2.33"
        },
        "latestPredictionTimestamp": {
          "description": "Latest prediction timestamp of a batch.",
          "format": "date-time",
          "type": [
            "string",
            "null"
          ],
          "x-versionadded": "v2.33"
        },
        "name": {
          "description": "Batch name.",
          "type": "string",
          "x-versionadded": "v2.33"
        }
      },
      "required": [
        "earliestPredictionTimestamp",
        "id",
        "latestPredictionTimestamp",
        "name"
      ],
      "type": "object"
    },
    "value": {
      "description": "Value of the metric in the bucket.",
      "type": [
        "number",
        "null"
      ],
      "x-versionadded": "v2.33"
    }
  },
  "required": [
    "batch",
    "value"
  ],
  "type": "object"
}

Service health stats of the deployment over a batch.

Properties¶

Name	Type	Required	Restrictions	Description
batch	Batch	true		Batch info.
value	number,null	true		Value of the metric in the bucket.

ServiceStatsOverBatchResponse

{
  "properties": {
    "buckets": {
      "description": "An array of buckets, representing service health stats of the deployment over selected batches.",
      "items": {
        "description": "Service health stats of the deployment over a batch.",
        "properties": {
          "batch": {
            "description": "Batch info.",
            "properties": {
              "earliestPredictionTimestamp": {
                "description": "Earliest prediction timestamp of a batch.",
                "format": "date-time",
                "type": "string",
                "x-versionadded": "v2.33"
              },
              "id": {
                "description": "Batch ID.",
                "type": "string",
                "x-versionadded": "v2.33"
              },
              "latestPredictionTimestamp": {
                "description": "Latest prediction timestamp of a batch.",
                "format": "date-time",
                "type": [
                  "string",
                  "null"
                ],
                "x-versionadded": "v2.33"
              },
              "name": {
                "description": "Batch name.",
                "type": "string",
                "x-versionadded": "v2.33"
              }
            },
            "required": [
              "earliestPredictionTimestamp",
              "id",
              "latestPredictionTimestamp",
              "name"
            ],
            "type": "object"
          },
          "value": {
            "description": "Value of the metric in the bucket.",
            "type": [
              "number",
              "null"
            ],
            "x-versionadded": "v2.33"
          }
        },
        "required": [
          "batch",
          "value"
        ],
        "type": "object"
      },
      "type": "array",
      "x-versionadded": "v2.33"
    },
    "metric": {
      "description": "Name of the metric requested.",
      "type": "string",
      "x-versionadded": "v2.33"
    },
    "segmentAttribute": {
      "description": "The name of the segment on which segment analysis is being performed.",
      "type": [
        "string",
        "null"
      ],
      "x-versionadded": "v2.20"
    },
    "segmentValue": {
      "default": "",
      "description": "The value of the `segmentAttribute` to segment on.",
      "type": [
        "string",
        "null"
      ],
      "x-versionadded": "v2.20"
    }
  },
  "required": [
    "buckets",
    "metric"
  ],
  "type": "object"
}

Properties¶

Name	Type	Required	Description
buckets	[ServiceStatsOverBatchBucket]	true	An array of buckets, representing service health stats of the deployment over selected batches.
metric	string	true	Name of the metric requested.
segmentAttribute	string,null	false	The name of the segment on which segment analysis is being performed.
segmentValue	string,null	false	The value of the `segmentAttribute` to segment on.

ServiceStatsOverTimeBucket

{
  "description": "Service health stats of the deployment over a time range.",
  "properties": {
    "modelId": {
      "description": "The id of the model for which metrics are being retrieved.",
      "type": "string",
      "x-versionadded": "v2.37"
    },
    "period": {
      "description": "An object with the keys \"start\" and \"end\" defining the period.",
      "properties": {
        "end": {
          "description": "End of the period to retrieve monitoring data, defaults to the next top of the hour. Note: this field only accepts top of the hour RFC3339 datetime strings, for example: ``2019-08-01T00:00:00Z``.",
          "format": "date-time",
          "type": [
            "string",
            "null"
          ]
        },
        "start": {
          "description": "Start of the period to retrieve monitoring data, defaults to 7 days ago from the end of the period. Note: this field only accepts top of the hour RFC3339 datetime strings, for example: ``2019-08-01T00:00:00Z``.",
          "format": "date-time",
          "type": [
            "string",
            "null"
          ]
        }
      },
      "type": "object"
    },
    "value": {
      "description": "Value of the metric in the bucket. Null if no value",
      "type": [
        "number",
        "null"
      ]
    }
  },
  "required": [
    "period",
    "value"
  ],
  "type": "object"
}

Service health stats of the deployment over a time range.

Properties¶

Name	Type	Required	Description
modelId	string	false	The id of the model for which metrics are being retrieved.
period	TimeRange	true	An object with the keys "start" and "end" defining the period.
value	number,null	true	Value of the metric in the bucket. Null if no value

ServiceStatsOverTimeResponse

{
  "properties": {
    "buckets": {
      "description": "An array of bucket, representing service health stats of the deployment over time.",
      "items": {
        "description": "Service health stats of the deployment over a time range.",
        "properties": {
          "modelId": {
            "description": "The id of the model for which metrics are being retrieved.",
            "type": "string",
            "x-versionadded": "v2.37"
          },
          "period": {
            "description": "An object with the keys \"start\" and \"end\" defining the period.",
            "properties": {
              "end": {
                "description": "End of the period to retrieve monitoring data, defaults to the next top of the hour. Note: this field only accepts top of the hour RFC3339 datetime strings, for example: ``2019-08-01T00:00:00Z``.",
                "format": "date-time",
                "type": [
                  "string",
                  "null"
                ]
              },
              "start": {
                "description": "Start of the period to retrieve monitoring data, defaults to 7 days ago from the end of the period. Note: this field only accepts top of the hour RFC3339 datetime strings, for example: ``2019-08-01T00:00:00Z``.",
                "format": "date-time",
                "type": [
                  "string",
                  "null"
                ]
              }
            },
            "type": "object"
          },
          "value": {
            "description": "Value of the metric in the bucket. Null if no value",
            "type": [
              "number",
              "null"
            ]
          }
        },
        "required": [
          "period",
          "value"
        ],
        "type": "object"
      },
      "type": "array"
    },
    "metric": {
      "description": "Name of the metric requested.",
      "type": "string"
    },
    "modelId": {
      "description": "The id of the model for which metrics are being retrieved. Deprecated, use modelId in each bucket instead.",
      "type": "string",
      "x-versiondeprecated": "v2.37"
    },
    "segmentAttribute": {
      "description": "The name of the segment on which segment analysis is being performed.",
      "type": [
        "string",
        "null"
      ],
      "x-versionadded": "v2.20"
    },
    "segmentValue": {
      "default": "",
      "description": "The value of the `segmentAttribute` to segment on.",
      "type": [
        "string",
        "null"
      ],
      "x-versionadded": "v2.20"
    },
    "summary": {
      "description": "Service health stats of the deployment over a time range.",
      "properties": {
        "modelId": {
          "description": "The id of the model for which metrics are being retrieved.",
          "type": "string",
          "x-versionadded": "v2.37"
        },
        "period": {
          "description": "An object with the keys \"start\" and \"end\" defining the period.",
          "properties": {
            "end": {
              "description": "End of the period to retrieve monitoring data, defaults to the next top of the hour. Note: this field only accepts top of the hour RFC3339 datetime strings, for example: ``2019-08-01T00:00:00Z``.",
              "format": "date-time",
              "type": [
                "string",
                "null"
              ]
            },
            "start": {
              "description": "Start of the period to retrieve monitoring data, defaults to 7 days ago from the end of the period. Note: this field only accepts top of the hour RFC3339 datetime strings, for example: ``2019-08-01T00:00:00Z``.",
              "format": "date-time",
              "type": [
                "string",
                "null"
              ]
            }
          },
          "type": "object"
        },
        "value": {
          "description": "Value of the metric in the bucket. Null if no value",
          "type": [
            "number",
            "null"
          ]
        }
      },
      "required": [
        "period",
        "value"
      ],
      "type": "object"
    }
  },
  "required": [
    "buckets",
    "metric"
  ],
  "type": "object"
}

Properties¶

Name	Type	Required	Description
buckets	[ServiceStatsOverTimeBucket]	true	An array of bucket, representing service health stats of the deployment over time.
metric	string	true	Name of the metric requested.
modelId	string	false	The id of the model for which metrics are being retrieved. Deprecated, use modelId in each bucket instead.
segmentAttribute	string,null	false	The name of the segment on which segment analysis is being performed.
segmentValue	string,null	false	The value of the `segmentAttribute` to segment on.
summary	ServiceStatsOverTimeBucket	false	Service health stats of the deployment over a time range.

TimeRange

{
  "description": "An object with the keys \"start\" and \"end\" defining the period.",
  "properties": {
    "end": {
      "description": "End of the period to retrieve monitoring data, defaults to the next top of the hour. Note: this field only accepts top of the hour RFC3339 datetime strings, for example: ``2019-08-01T00:00:00Z``.",
      "format": "date-time",
      "type": [
        "string",
        "null"
      ]
    },
    "start": {
      "description": "Start of the period to retrieve monitoring data, defaults to 7 days ago from the end of the period. Note: this field only accepts top of the hour RFC3339 datetime strings, for example: ``2019-08-01T00:00:00Z``.",
      "format": "date-time",
      "type": [
        "string",
        "null"
      ]
    }
  },
  "type": "object"
}

An object with the keys "start" and "end" defining the period.

Properties¶

Name	Type	Required	Restrictions	Description
end	string,null(date-time)	false		End of the period to retrieve monitoring data, defaults to the next top of the hour. Note: this field only accepts top of the hour RFC3339 datetime strings, for example: `2019-08-01T00:00:00Z`.
start	string,null(date-time)	false		Start of the period to retrieve monitoring data, defaults to 7 days ago from the end of the period. Note: this field only accepts top of the hour RFC3339 datetime strings, for example: `2019-08-01T00:00:00Z`.