[Integration]Service SLO dashboard

[YANG-DB]

What is the integration resource ?

We need a way for presenting a service SLO/SLI dashboard for understanding the business indications of the service performance and possible degradation

What is the integration source protocol ?

SLO are business indicators defined by the Customer and are highly relevant for the customer's specific environment and requirements.

It may be possible that the SLO measures will be taken based on one of the following features:

• service metrics
• service traces
• service logs

These signals may be residing on OpenSearch indices directly or on a remote datasource such as Prometheus or S3.

What is the integration target protocol ?

Eventually Data will be aggregated into a single index which will reflect the hourly/daily/timely summary of the different services SLO/SLI indicators that will be later displayed into the SLO dashboard suggested below .

Open Question:

• where will the SLO rules be defined ?
• how would the user be able to update these rules ?
• can alerts be directly ingested into opensearch ?

Would you be using an ingestion pipeline ?

Data may arrive via OTEL collector's

• data-prepper
• opensearch exporter
• jaeger

Via Flint's ingestion batch processing from S3.
Via Prometheus Metrics data using PPL datasource scheduled queries

Which Dashboards would you be using ?

Do you have any additional context?

• multiwindow-multi-burn-rate-alerts
• OpenSLO concepts
• Utilize Monitoring and Alerts notifications