Monitoring Spec Example

The following model exemplifies a configuration intended to be created by Vertical Slicer and posted to an endpoint from MDA:

{
    "business_id": "business_id",
    "topic": "topic_test",
    "monitoring_endpoint": "url_to_fetch",
    "tenant_id": "operator_id",
    "context_ids": [
        {
            "resource_id": "resource_id",
            "network_slice_id: "slice_id",
            "parent_id": "parent_resource_id"
        }
    ]
    "reference_id": "reference_id",
    "timestamp_start": "2021-04-06T17:36:00",
    "timestamp_end": "2021-05-28T16:27:00",
    "metrics": [
        {
            "metric_name": "cpu_utilization",
            "metric_type": "float",
            "aggregation_method": null,
            "step": "30s",
            "step_aggregation": null
        },
        {
            "metric_name": "disk_usage",
            "metric_type": "float",
            "aggregation_method": null,
            "step": "2m",
            "step_aggregation": null
        },
        {
            "metric_name": "ram_utilization",
            "metric_type": "float",
            "aggregation_method": "sum",
            "step": "2m",
            "step_aggregation": "5m"
        }
    ]
}

Resource Monitoring information

Parameter	Type	Description
business_id	String
topic	String
monitoring_endpoint	String
resource_id	String	Unique ID of resource for which we are collecting monitoring data.
network_slice_id	String
parent_id	String
reference_id	String	Unique ID of entity (e.g., SLA) for whom we are collecting monitoring data.
metric_name	String	Name of metric being monitored.
metric_value	String	Value of metric being monitored.
timestamp	Datetime	The time at which the metric was measured.

The resourceID represents the resource for which we are collecting monitoring data. If the resource used to be part of one SLA and is now part of another SLA, it may make sense to use a distinct resourceID for the different epochs, so that data from one epoch is not used to influence decisions in another epoch. The referenceID might be the SLA identifier to which this resource is dedicated and is used to look up all relevant monitoring data for that reference. The monitoring data provided to postMonitoringData() should be a list of such structures, with one entry per metric being recorded.