AWS has launched an enhanced Custom Event Bus for Amazon EventBridge that allows a single event bus to be shared across multiple AWS accounts, addressing infrastructure fragmentation that emerges when event-driven systems scale beyond one team. The new model supports up to 10,000 subscribers by default, with options to request higher quotas, and introduces native event ordering, retention up to 365 days, replay capabilities, content-based deduplication, and a redesigned pricing structure. AWS says the update tackles an architectural challenge common in enterprise environments where separate teams operate isolated accounts but need coordinated event handling.
The enhanced bus can be shared across AWS accounts using AWS Resource Access Manager, letting platform teams establish a central event backbone while individual application teams publish and subscribe from their own accounts. Events can be retained for a configurable period between one and 365 days, enabling consumers to start from an earlier point in event history or recover after failures. AWS also added synchronous invocation for targets like Lambda, content-based deduplication that collapses matching events received within a five-minute window, and support for JSONata transformations alongside deserialization of Avro and Protocol Buffers payloads into JSON for filtering. The company introduced a new Subscriber resource that combines filtering, target configuration, retries, and dead-letter destinations into a single consumer definition with variable start times for onboarding or replay scenarios.
According to the report, organizations commonly use multiple accounts to isolate teams and applications, but this leads to EventBridge buses being connected through cross-account rules and forwarding configurations that become harder to govern as consumer numbers grow. AWS states that forwarding the same event through multiple buses can introduce additional processing and routing costs. The company says the enhanced bus supports both conventional asynchronous event consumption and workloads where ordering and processing confirmation are important, though content-based deduplication "does not remove the need for application-level idempotency, particularly where events are replayed or processed across regions."
The shift changes governance as much as technology. Rather than platform teams configuring every cross-account connection, they can now set boundaries around the central bus while individual application teams manage their own subscriptions, filtering, targets, retries, and failure destinations—creating separation between platform-level oversight and application-level event consumption that reduces infrastructure work needed to onboard new consumers. AWS has also replaced the per-hop cost model with ingress and egress throughput pricing, where publishers pay for events ingested and subscribers pay for events delivered, aligning charges more closely with data volume entering and leaving the central bus rather than intermediate routing steps. The report notes that the enhanced model moves into territory traditionally occupied by enterprise event brokers and streaming platforms like Apache Kafka, but EventBridge remains a managed, serverless event-routing service rather than a direct replacement for Kafka in every streaming architecture.
AWS isn't retiring the existing EventBridge model immediately—current custom buses continue operating unchanged and are now labeled "Custom event bus-classic," while the enhanced bus launches as a separate resource, allowing organizations to adopt the new architecture incrementally without redesigning existing applications. The enhanced Custom Event Bus is initially available across selected AWS Regions in North America, Europe, and Asia Pacific, and can be created through the AWS Management Console, CLI, or EventBridge APIs. For enterprises running multi-account AWS estates, the move signals a bet that centralized event infrastructure can coexist with decentralized team autonomy, though the real test will be whether governance models catch up to the technical capabilities. The challenge isn't just whether platform teams can provision shared buses—it's whether organizations can agree on who owns event contracts when dozens of teams depend on the same stream.