Security Data Works

Component reference

Trino — federation breadth

Engine-anchored architecture for multi-region, multi-platform SOCs that cannot, or should not, centralize all data into one store. Federate first, then query. Standard SQL throughout, so detection content stays portable.

3.6×

Faster than the schema-on-read SIEM foil (OpenSearch, 2.854s) on the same Zeek workload at 0.795s average query (single host, Tier B), with federation reach the schema-on-read SIEM tier doesn't offer. Iceberg, Postgres, cloud DWs, and Kafka queried in one SQL plane, without copying data into a central store.

The pipeline

  1. Iceberg

    Lakehouse

    Long-term retention · hunting

  2. Postgres / DWs

    Operational stores

    Snowflake, BigQuery, Postgres

  3. Engine

    Trino federation plane

    Single SQL surface, all sources

  4. Kafka

    Streams

    Near-real-time joins

  5. Serve

    SOC + analytics

    Investigation, ad-hoc hunt

What composes, what’s brittle

  • No central store required. Data stays where it sits; query reaches across.
  • Why Trino. Mature connector ecosystem; standard SQL; horizontal scale-out.
  • Best fit. Multi-region, data-residency, cross-jurisdiction SOCs.
  • Trade-off. Slower than single-engine OLAP (0.795s vs. ClickHouse native 0.061s), and the highest join tax the SDW join bench measured: joined-over-flat ratio 4.35× on Trino where StarRocks and both ClickHouse arms sat near 1.8× (single host, Tier B).
  • What survives. Standard SQL detections portable across engines.
  • What's distinctive. Federation reach the schema-on-read SIEM tier doesn't offer.

Sources: Methodology: published spec, reproducible on equivalent hardware · Trino documentation

See how the pattern lands on your workload.

The matrix scoring that justified each reference architecture's tool choices is the paid deliverable. The benchmark behind it is public — reproduce it on your own workload, then book a call to scope the work.