Understanding synchronous updates in Reltio Lightspeed
Learn about the synchronous write mode in Reltio Lightspeed™ Data Delivery Network (Reltio Lightspeed) that guarantees read-after-write consistency for mission-critical APIs.
If your operational workflows rely on immediate access to newly written data, the Reltio Lightspeed™ Data Delivery Network (Reltio Lightspeed) provides an optional synchronous update mode to guarantee read-after-write consistency.
What is it?
This feature ensures that updates to entities and relationships are synchronously written to the Lightspeed Predicate Index. When you enable this feature, write operations wait until the index is updated before returning a successful response.
Who is it for?
This content is curated for these Reltio user roles defined in topic About roles:
Reltio Lightspeed is designed for:
- Developers who need a quick way to fetch specific data.
- Data and IT teams managing high-performance data queries.
- Businesses that rely on real-time data for analytics, AI, and customer interactions.
Developer
Reltio Configurator
System Administrator
Why would I use it?
With strong consistency, your Reltio Lightspeed reads always reflect the latest writes. This removes the need for fallback logic, retries, or delays in your orchestration flows.
When and where would I use it?
- Customer profile creation followed by activation or validation.
- Immediate response validation in fraud detection services.
- Transaction flows involving create + lookup in the same session.
How does it support business goals?
Enabling this feature allows your API-first integrations to operate with confidence, improving user experience and reducing integration complexity without requiring deep understanding of Reltio’s internal queues or index propagation patterns.
For details on configuration fields and override rules, see Configuration options for synchronous consistency in Reltio Lightspeed.
How synchronous write behavior works
When you enable synchronous updates in Reltio Lightspeed, the platform uses a coordinated two-stage process to ensure that new or updated records are queryable immediately after they're written.
This process guarantees read-after-write consistency by updating both:
- The core transactional store: First, your entity or relationship is written to Reltio’s primary data store. This is either Amazon DynamoDB if you're deployed on AWS, or Google Cloud Spanner if you're on GCP. These distributed database systems are designed to handle high-throughput, low-latency writes while maintaining durability and consistency.
- The Lightspeed Predicate Index: After writing to the core, the system synchronously updates a dedicated index that powers fast query access via the Predicate API. This index is backed by a distributed key-value store optimized for real-time lookups.
Only after both steps are successfully completed does the API return a success response. This atomic behavior ensures that your data is fully committed and immediately available for querying — without requiring retry logic, fallback reads, or delay in downstream processes.