Use case · DuckDB
Analytics & indexers
If you already speak DuckDB, StreamSync lets you push analytical work out to a shard network — and keep it fast as Solana state grows.
The problem
Indexing Solana into an analytical store works until the store becomes the bottleneck. One warehouse gets big, slow, and expensive; scaling it is a project, and every new dashboard query competes for the same resources.
How StreamSync helps
- Sharded DuckDB. A hash ring partitions state across operator clusters. Each operator runs embedded DuckDB over its partition and merges sub-query results locally.
- Analytical SQL as a first-class surface. Anything expressible in SQL is fair game — GROUP BYs across token accounts, holder distributions, program activity over time.
- Archive nodes for history. The archive specialization (high storage) serves deep historical shards so long-range queries stay answerable.
- Pay per successful query. The billing model rewards fast, correct answers — not warehouse uptime you provision and babysit.
Getting started
Read how the DuckDB sharding and race pipeline fit together, then wire the gateway. See the deep-dive comparison in the blog and the rest of the use cases. Part of Cryptuon.
FAQ
We already run DuckDB. Why StreamSync?
StreamSync’s query layer is DuckDB, sharded across operator clusters by a hash ring. You push analytical SQL out to the network instead of scaling one warehouse, and sub-query results merge locally so queries stay fast at multi-billion-row scale.
What about historical and archival data?
A dedicated archive node class (high storage) serves deep historical shards and backfills, so long-range analytical queries do not fall off a performance cliff.