Sui Introduces gRPC Streaming for Real-Time Blockchain Indexing

The Sui blockchain has rolled out gRPC streaming as a core data source for its indexing infrastructure, marking a shift toward real-time data ingestion. This update allows developers to receive finalized checkpoint data with minimal delay, which is especially important for applications that need to respond immediately once blockchain data is confirmed. By prioritizing streaming as a primary ingestion method, Sui aims to support faster reactions and more responsive tooling across its ecosystem.

At the same time, the platform has not abandoned existing approaches. Instead, it blends streaming with traditional polling mechanisms to ensure data completeness and operational stability. This hybrid approach preserves compatibility with custom indexers that are already live on Sui, allowing teams to enhance performance without redesigning their systems from scratch.

Streaming Eliminates Polling Delays

The introduction of gRPC streaming significantly changes how indexers consume blockchain data. Rather than repeatedly querying full nodes at fixed intervals, indexers now receive checkpoint updates pushed directly from nodes as soon as finalization occurs. This push-based delivery removes the delays that were inherent in polling-based models, where data freshness depended heavily on fetch frequency and timing configuration.

As a result, applications that are sensitive to latency can operate closer to real time. Monitoring tools, alerting systems, and live analytics platforms benefit in particular, since they no longer need to balance infrastructure costs against aggressive polling schedules. The streamlined setup also reduces operational complexity by eliminating the need for constant tuning of polling intervals.

Hybrid Model Protects Data Integrity

While streaming improves speed, Sui pairs it with mandatory polling fallbacks to address the limitations of long-lived connections. Because a streaming connection only provides data from the moment it is established, historical checkpoints still require retrieval through polling-based sources.

This hybrid pattern is already in use within Sui’s General-Purpose Indexer. In that setup, streaming serves as the primary ingestion path for new data, while polling remains available to backfill historical checkpoints and support recovery scenarios. If a streaming connection drops or an indexer restarts, polling enables the system to resume from the last processed checkpoint before switching back to streaming.

This design ensures that indexers remain both current and complete, even in the face of network interruptions or service disruptions. Developers gain the advantages of low-latency updates without risking data gaps or inconsistencies.

Framework Supports Flexible Adoption

Sui’s Custom Indexing Framework plays a key role in enabling this transition. The framework clearly separates checkpoint processing logic from data ingestion methods. Indexers interact with checkpoints through a unified interface, regardless of whether the data originates from gRPC streams or HTTP-based polling.

Because of this abstraction, teams can modify how they ingest data without rewriting their core processing logic. Streaming can be introduced incrementally, depending on the needs of each workload. Applications that demand real-time responsiveness can adopt streaming immediately, while batch-oriented or offline systems can continue relying solely on polling.

Minimal Configuration, Stable Logic

For existing indexers built on the official framework, enabling gRPC streaming requires only minor configuration changes. Developers can add a streaming endpoint alongside their existing remote storage settings, while leaving checkpoint processing logic untouched. The framework automatically coordinates between streaming and polling sources to maintain a consistent view of the blockchain.

Overall, Sui’s combined streaming and polling architecture delivers faster access to finalized checkpoints, stronger resilience, and a smooth upgrade path. This approach positions the network to better support real-time applications while maintaining the reliability required for long-term data indexing.