Skip to main content

Deployment Options

InfraSage offers two deployment models. Both run the same platform — they differ in who manages the infrastructure.

Comparison​

Cloud (InfraSage Hosted)Self-Hosted
InfrastructureFully managed by InfraSageRuns in your cloud account
SetupSign up and start in minutesDeploy via provided Kubernetes manifests
Data residencyInfraSage-managed regionsFully within your environment
ScalingAutomaticConfigured via HPA and scale profiles
UpgradesAutomaticApplied via rolling manifest updates
SLAUp to 99.99% (Enterprise plan)Dependent on your infrastructure
Best forMost teams and POCsRegulated industries, air-gapped environments

Cloud (InfraSage Hosted)​

The hosted service at console.infrasage.dev requires no infrastructure to manage. InfraSage operates and scales all platform components on your behalf.

Getting Started​

  1. Sign up at console.infrasage.dev/register
  2. Create an API key under Settings → API Keys
  3. Point your telemetry sources at your ingestion endpoint

Your ingestion endpoint, API keys, usage metrics, and team settings are all managed through the console.

When to Choose Cloud​

  • You want to start quickly without infrastructure overhead
  • Your team lacks dedicated platform/infra resources
  • You're running a proof of concept
  • You want InfraSage-managed SLA guarantees

Self-Hosted​

Self-hosted deployments run the full InfraSage stack inside your Kubernetes cluster. You control the network, storage, and compute — InfraSage provides the artifacts.

Prerequisites​

  • Commercial license from InfraSage ([email protected])
  • Kubernetes 1.25+ cluster
  • Access to the InfraSage private artifact registry (provided with license)

Architecture​

The self-hosted deployment uses the same microservices as the cloud offering:

External Sources
      │
      ▼
Ingestion Gateway (8080)
      │
      ▼ Kafka (Redpanda)
      │
      ▼
Telemetry Operator (8081)
      │
      ▼ SQL
      │
ClickHouse (time-series DB)
      │
      ▼ SQL
      │
AIops Engine (8080/9093)
      │
      ▼
Integrations (Slack, PagerDuty, Jira, Teams, Webhooks, ...)

Scaling​

InfraSage components scale horizontally. The Ingestion Gateway, Telemetry Operator, and AIops Engine all support Kubernetes HPA:

# Manual scale
kubectl scale deployment ingestion-gateway -n infrasage --replicas=5

# Autoscale
kubectl get hpa -n infrasage
# NAME                       MINPODS   MAXPODS   REPLICAS   CPU
# ingestion-gateway          2         10        3          45%

See Scale Profiles for recommended sizing across small, medium, and large environments.

Accessing Services (Port Forward for Admin Access)​

kubectl port-forward svc/prometheus 9999:9090 -n infrasage &
kubectl port-forward svc/grafana 3000:3000 -n infrasage &
kubectl port-forward svc/ingestion-gateway 8080:8080 -n infrasage &

Setting Up Ingress​

# Traefik is included in K3S by default; any ingress controller works
kubectl apply -f - <<EOF
apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
  name: infrasage-ingress
  namespace: infrasage
spec:
  rules:
  - host: infrasage.mycompany.com
    http:
      paths:
      - path: /
        pathType: Prefix
        backend:
          service:
            name: ingestion-gateway
            port:
              number: 8080
EOF

When to Choose Self-Hosted​

  • Regulatory or compliance requirements mandate data residency (HIPAA, FedRAMP, PCI-DSS)
  • Your environment is air-gapped or behind a private network
  • You need custom storage backends or retention policies beyond plan limits
  • You have a large existing Kubernetes investment and prefer to operate the stack yourself

Choosing a Region (Cloud)​

InfraSage cloud is available in multiple regions. Select your region during signup or via Settings → Region in the console. Ingestion endpoints are regionally scoped to minimize latency from your services.

Security Considerations​

Regardless of deployment model, review the Security section before going to production:

  • API key scoping and rotation
  • RBAC configuration
  • Audit log retention
  • TLS on all ingestion endpoints