// cloudflare deployment patterns

Six ways to deploy Cloudflare, one diagram each

Common deployment topologies, from a standard front door for an AWS origin to multi-region load balancing, private access, API security, and log delivery into Splunk and Microsoft Sentinel. One line and one diagram per pattern.

A

Standard app to an AWS ALB origin

Cloudflare sits in front of a single AWS origin, handling DNS, TLS, DDoS, WAF and caching at the edge, with the origin locked down so clients cannot bypass Cloudflare and reach the ALB directly.

Diagram A: Standard app to an AWS ALB origin
B

Multi-region with Cloudflare Load Balancing

Cloudflare steers traffic across health-checked application pools in two AWS regions and fails over automatically, provided the health monitors reflect real ability to serve rather than a non-critical dependency.

Diagram B: Multi-region with Cloudflare Load Balancing
C

Private admin via Tunnel and Access

An admin application with no public ingress is reached through an outbound-only cloudflared tunnel, with Cloudflare Access enforcing identity, group, MFA and device posture before any connection is allowed.

Diagram C: Private admin via Tunnel and Access
D

API security deployment

Edge controls (DDoS, WAF, rate limiting, bot signals and API Shield with mTLS, JWT and schema validation) protect an AWS API origin, while the application still enforces authorisation, because a valid JWT proves the token, not the user’s right to the object.

Diagram D: API security deployment
E

Cloudflare to Splunk

Cloudflare datasets stream by direct Logpush into Splunk HEC as JSON, landing in an index that feeds dashboards, correlation and detections.

Diagram E: Cloudflare to Splunk
F

Cloudflare to Microsoft Sentinel

Cloudflare Logpush writes to an Azure Blob container that the Cloudflare connector ingests into Sentinel, where the logs drive analytics rules, workbooks, hunting and automation playbooks.

Diagram F: Cloudflare to Microsoft Sentinel