AWS Security ChangesHomeSearch

AWS wellarchitected: Rewrite ADVREL08-BP01 to cover ad-verification failover

Service: wellarchitected · 2026-11-11 · Documentation low

File: wellarchitected/latest/video-streaming-advertising-lens/advrel08-bp01.md

Summary

Replaces the previous privacy-preserving fault-tolerance guidance (AWS Clean Rooms, differential privacy, Nitro Enclaves) with new reliability guidance on redundant ad-verification providers, health-based routing, circuit breakers, SQS dead-letter queues and DynamoDB deduplication. Retitles the best practice, adds desired outcome, anti-patterns, benefits, implementation steps, risk level, related practices/documents/services, and updates breadcrumb labels.

Security assessment

The diff is a reliability/content rewrite: it swaps privacy/encryption-oriented text (Clean Rooms replicated privacy policies, cross-Region KMS, encrypted retry queues, fail-closed circuit breakers) for availability-oriented content about third-party ad-verification providers, health checks, circuit breakers, dead-letter queues and event-ID deduplication. The quoted line and surrounding additions concern measurement continuity and duplicate-count avoidance, not a vulnerability, credential, IAM, encryption or public-exposure issue; no CVE or advisory is referenced.

Evidence

+Implement redundant ad-verification systems with automated failover. Use multiple verification providers and automated monitoring for reliable advertising measurement and validation.

Diff

diff --git a/wellarchitected/latest/video-streaming-advertising-lens/advrel08-bp01.md b/wellarchitected/latest/video-streaming-advertising-lens/advrel08-bp01.md
index e69e5d7..4521732 100644
--- a//wellarchitected/latest/video-streaming-advertising-lens/advrel08-bp01.md
+++ b//wellarchitected/latest/video-streaming-advertising-lens/advrel08-bp01.md
@@ -7 +7 @@
-Implementation guidanceKey AWS services:Resources
+Implementation guidanceResources
@@ -9 +9 @@ Implementation guidanceKey AWS services:Resources
-# ADVREL08-BP01 Design resilient architectures with privacy-preserving fault tolerance
+# ADVREL08-BP01 Implement redundant ad verification systems with automated failover mechanisms
@@ -11 +11 @@ Implementation guidanceKey AWS services:Resources
-Build resilient architectures that maintain data privacy in multi-party collaborations, focusing on fault-tolerant AWS Clean Rooms deployment, encrypted failover mechanisms, and privacy-preserving disaster recovery procedures. 
+Implement redundant ad-verification systems with automated failover. Use multiple verification providers and automated monitoring for reliable advertising measurement and validation. 
@@ -13 +13,27 @@ Build resilient architectures that maintain data privacy in multi-party collabor
-## Implementation guidance
+**Desired outcome:**
+
+  * Multiple verification providers operate in parallel, and failover to backup providers is automatic. 
+
+  * Measurement continues uninterrupted when individual verification providers experience degradation. 
+
+  * Circuit breakers guard against degraded third-party services blocking measurement pipelines. 
+
+
+
+
+**Common anti-patterns:**
+
+  * Relying on a single verification provider, so its degradation halts all measurement. 
+
+  * No circuit breaker on third-party verification calls, so a degraded provider can block the measurement pipeline. 
+
+  * Processing measurement events without deduplication, so retries inflate counts. 
+
+  * No health-based routing or automated failover between providers, so failover requires manual intervention. 
+
+
+
+
+**Benefits of establishing this best practice:**
+
+  * Multiple providers operate in parallel with automatic failover, so measurement continues during provider degradation. 
@@ -15 +41 @@ Build resilient architectures that maintain data privacy in multi-party collabor
-  * Deploy AWS Clean Rooms across multiple Regions with replicated privacy policies, differential privacy budgets, and encrypted collaboration configurations to facilitate continuous privacy-protected analytics during regional outages. 
+  * Circuit breakers keep degraded third-party services from blocking measurement pipelines. 
@@ -17 +43 @@ Build resilient architectures that maintain data privacy in multi-party collabor
-  * Configure automatic failover for AWS Clean Rooms and Nitro Enclaves with cross-Region KMS key access, synchronized IAM roles, and validated privacy control restoration to maintain cryptographic isolation and data protection during service transitions. 
+  * DynamoDB deduplication avoids duplicate counting when events are retried. 
@@ -19 +44,0 @@ Build resilient architectures that maintain data privacy in multi-party collabor
-  * Implement privacy-aware error handling for data matching with encrypted retry queues, failed operation logging that preserves anonymity, and automatic termination of computations that cannot maintain privacy guarantees during processing errors. 
@@ -21 +45,0 @@ Build resilient architectures that maintain data privacy in multi-party collabor
-  * Deploy circuit breakers with privacy validation that fail-closed when privacy controls cannot be verified, monitor differential privacy budget exhaustion, and halt operations when cryptographic attestation fails in dependent services. 
@@ -23 +46,0 @@ Build resilient architectures that maintain data privacy in multi-party collabor
-  * Monitor AWS Clean Rooms privacy metrics including query result threshold compliance, privacy budget consumption rates, unauthorized access attempts, and cryptographic operation health with automated alerts for privacy policy violations. 
@@ -25 +48 @@ Build resilient architectures that maintain data privacy in multi-party collabor
-  * Use encrypted dead-letter queues for failed matching operations with privacy context preservation, secure purging policies for expired operations, and manual review processes that maintain data anonymization during failure analysis. 
+**Level of risk exposed if this best practice is not established:** High 
@@ -27 +50 @@ Build resilient architectures that maintain data privacy in multi-party collabor
-  * Automate backup of privacy-protected datasets with cross-Region encrypted replication, privacy policy version control, differential privacy state preservation, and recovery procedures that validate privacy controls before data restoration. 
+## Implementation guidance
@@ -28,0 +52 @@ Build resilient architectures that maintain data privacy in multi-party collabor
+IAS, DoubleVerify, and MOAT verify advertising measurement for advertisers. Because their availability is outside your control, a single-provider design ties your measurements to their availability. MRC accreditation and advertiser billing both depend on accurate measurement. 
@@ -29,0 +54 @@ Build resilient architectures that maintain data privacy in multi-party collabor
+Two providers running in parallel can measure the same event. Agreement between them cross-validates the numbers, and disagreement signals a data quality issue worth investigating. Each added provider has its own integration cost. Route each request to a provider based on current health. 
@@ -30,0 +56 @@ Build resilient architectures that maintain data privacy in multi-party collabor
+A measurement can drop without producing an error with no alarm to signal that loss. The loss then appears later during reconciliation, by which point Amazon Kinesis retention has expired and the original events are gone. Send failed measurements to an Amazon SQS dead-letter queue for deferred reprocessing. 
@@ -32 +58 @@ Build resilient architectures that maintain data privacy in multi-party collabor
-## Key AWS services:
+A retried event can be counted twice if nothing tracks prior processing, and duplicate counting corrupts billing numbers. Store each event's ID in DynamoDB so retries can be identified and skipped. Circuit breaker activations feed SLA tracking and partner reporting. 
@@ -34 +60 @@ Build resilient architectures that maintain data privacy in multi-party collabor
-  * AWS Clean Rooms 
+### Implementation steps
@@ -36 +62 @@ Build resilient architectures that maintain data privacy in multi-party collabor
-  * Amazon Route 53 
+  1. **Integrate multiple verification providers with health-based routing:** Configure Amazon API Gateway or an application-level routing layer that routes measurement requests to verification providers based on real-time health checks. Define primary and fallback providers per measurement type (viewability, click verification, conversion tracking). Implement automated health checks that test each provider every 60 seconds. 
@@ -38 +64 @@ Build resilient architectures that maintain data privacy in multi-party collabor
-  * AWS Auto Scaling 
+  2. **Implement circuit breakers for third-party verification calls:** Deploy circuit breakers on all calls to external verification providers. Set open threshold at 30% failure rate over 60 seconds. Configure fallback to: use alternative provider, cache last-known-good measurement data, or record event for later verification. Log all circuit breaker activations for SLA tracking and partner reporting. 
@@ -40 +66 @@ Build resilient architectures that maintain data privacy in multi-party collabor
-  * Amazon EventBridge 
+  3. **Deploy DynamoDB for measurement deduplication:** Store measurement event IDs in DynamoDB with a 24-hour TTL. Before processing any measurement event, check the DynamoDB table for a previous processing record. This avoids duplicate counting when events are retried after failures. Use DynamoDB conditional writes to make deduplication atomic. 
@@ -47 +73,27 @@ Build resilient architectures that maintain data privacy in multi-party collabor
-  * [Reliablity Design principles](https://docs.aws.amazon.com/wellarchitected/latest/reliability-pillar/design-principles.html)
+**Related best practices:**
+
+  * [ADVREL04-BP02 Architect defensively against failures](./advrel04-bp02.html)
+
+  * [ADVREL08-BP02 Establish reliable data collection and validation pipelines for measurement accuracy](./advrel08-bp02.html)
+
+
+
+
+**Related documents:**
+
+  * [Implementing health checks - Amazon Builders' Library](https://aws.amazon.com/builders-library/implementing-health-checks/)
+
+  * [Making retries safe with idempotent APIs - Amazon Builders' Library](https://aws.amazon.com/builders-library/making-retries-safe-with-idempotent-apis/)
+
+
+
+
+**Related services:**
+
+  * [Amazon CloudWatch](https://aws.amazon.com/cloudwatch/)
+
+  * [AWS Lambda](https://aws.amazon.com/lambda/)
+
+  * [Amazon EventBridge](https://aws.amazon.com/eventbridge/)
+
+  * [AWS Step Functions](https://aws.amazon.com/step-functions/)
@@ -49 +101 @@ Build resilient architectures that maintain data privacy in multi-party collabor
-  * [Disaster recovery best practices](https://docs.aws.amazon.com/clean-rooms/latest/userguide/disaster-recovery-resiliency.html)
+  * [Amazon DynamoDB](https://aws.amazon.com/dynamodb/)
@@ -60 +112 @@ To use the Amazon Web Services Documentation, Javascript must be enabled. Please
-Privacy
+Ad measurement and verification reliability
@@ -62 +114 @@ Privacy
-ADVREL08-BP02 Maintain data consistency and availability across collaboration workflows
+ADVREL08-BP02 Establish reliable data collection and validation pipelines for measurement accuracy