Mission
Most applications that operate at the edge — desktop software, field devices, IoT sensors, mobile apps — face the same unsolved problem: how do you reliably get data from the edge to the cloud without building a custom replication pipeline?
The standard answers are either "stream everything over HTTP in real time" (brittle, requires constant connectivity) or "write your own CDC and sync logic" (months of engineering, perpetual maintenance). SyncLite offers a third path: embed a proven database engine at the edge, let SyncLite capture every transaction as a compact binary log, and let the Consolidator handle delivery — with exactly-once semantics, offline resilience, and zero custom code.
The goal is to make data synchronization as unremarkable as HTTP — something you configure in minutes, not build over months.
Offline First, Always
Edge devices operate fully without connectivity. Data accumulates locally and syncs the moment a path exists.
Exactly Once, Every Time
Transactional log capture at the source guarantees no duplicates and no data loss — regardless of network conditions.
Open & Composable
Apache 2.0. No lock-in to any cloud, staging system, or destination. Swap components without touching your app.
Low Code by Design
Standard JDBC, a simple HTTP API, and a web UI. No new query language. No proprietary SDK to learn.
Author
Patent
🔏 System and Method for Handling Data Consolidation from an Edge Device to a Target Database
SyncLite's core architecture is protected by granted patents and published international filings that cover transaction-log capture on edge devices, staged transfer, and exactly-once consolidation into a target database.
The patented flow describes: application transactions on edge devices, wrapper-based transaction log capture, storage in device-specific remote directories, and a consolidator/replicator pipeline that reads logs, produces change logs, schedules consolidation, and applies schema plus data into destination databases.
Patent references:
- US Patent: US 12,505,123 B2 — Granted on 23-12-2025
- India Patent Office: 429657 — Granted on 21-04-2023
- PCT Publication: WO/2023/111765 — Published on 22-06-2023
In practical terms, this covers offline-resilient edge logging, staged synchronization, and deterministic transaction replay into centralized targets with high integrity and low operational complexity.
Milestones
Project Started
SyncLite began as an internal tool to solve a real edge-to-cloud synchronization problem — before any open-source release.
Patent Filed
The core synchronization architecture was filed for patent protection as the platform matured into a production-ready system.
Open Source Release
SyncLite was released publicly on GitHub under Apache License 2.0. Full platform — Logger, Consolidator, DBReader, QReader, Job Monitor, Validator — open to the community.
v2024.10.02 Released
Stable release with multi-destination consolidation, 16 device types, SyncLiteStore / SyncLiteStream APIs, HTTP/JSON server, and full Docker support.
Contact
Need help, want to contribute, or want to discuss your SyncLite deployment?
Getting Started Guide
Step-by-step setup instructions for local deployment and first sync.
Open Guide →Useful Links
Everything you need to explore, use, and contribute to SyncLite: