Avisi Cloud Observability.
Avisi Cloud Observability product overview.
Avisi Cloud Observability is an offering included within our AME Kubernetes product and is aimed at helping developers get the best performance out of their applications running on Kubernetes.
Part of the Application Observability is our integrated observabililty solution using:
- Prometheus for metrics collection
- Loki for log events
- Grafana for visualizing the data
- Dynatrace (enterprise only, currently in beta) as a fully fledged application performance management tool
We heavily build upon open standards and common best-in-class tooling that can be run anywhere for our Performance & Observability tooling.
Our focus is on integrating best-in-class existing tooling with our Kubernetes offering, giving you fully access and functionality. The differences between the SLA types (basic, premium and enterprise) are only in retention periods of the data collected.
|Monitoring||In cluster Prometheus for collecting metrics, with external long term storage.||7 days||30 days||90 days|
|Logging||Log collection and storage using Loki, with external long term storage.||7 days||30 days||90 days|
|Alerting||Alertmanager for routing Prometheus & Loki Alerts to external tools|
|HA Grafana access||Grafana access for your organisation, for viewing metrics and logging data.|
|Application Performance Monitoring*||Optional access to Dynatrace||-||Infrastructure only|
|Audit logging||Audit logging of cluster activity|
* Note that Application Performance Monitoring is currently only available in a closed beta form. If you are interested in making use of Dynatrace APM, please contact Avisi Cloud.
Clusters come with the Prometheus operator installed and a Prometheus instance configured to forward any metrics data to long term storage. All standard USE metrics of your nodes and pods are automatically available to you.
Log events within clusters are collected and forwarded to Avisi Cloud’s loki stack.
Use alertmanager to route your Prometheus alerts to any external source such as Slack, Opsgenie or custom webhooks.
- Alerting is currently only possible on metric data from Prometheus.