Real-time Iceberg
Without Pipelines
Stop waiting on batches. With its unique zero-copy approach, Streambased turns your real-time Kafka data into your single Iceberg unified view in milliseconds, not hours.
New Data - Trapped in Kafka
Every event starts its life in Kafka. But for most teams, that’s where it stops.Before it can reach dashboards, models, or analysts, data must be copied, batched, and re-written elsewhere. What this means:
• New data sits idle until the next pipeline run.
• Analysts wait while the system catches up.
• Kafka stays fresh but everything downstream lags behind.
Streambased begins here: the moment data lands.
Copying Data
Creates Friction
To make Kafka data queryable, teams build copy pipelines into Iceberg, but that approach always hits limits:
• Freshness lag: to write efficient Iceberg files, pipelines buffer events into big chunks - so data isn’t queryable until the batch closes.
• Maintenance churn: frequent small writes lead to lots of tiny files; you pay with compaction and snapshot expiration loops.
• Two copies mean no single truth: every copy risks data skew and drift across systems due to retries, replays and schema mismatches.
• Partition mismatch: Kafka partitions for throughput; Iceberg partitions for pruning - optimize for one and you harm the other.
Streambased’s Unique View-Based Approach
Streambased projects Kafka into Iceberg at query time - no background copy. Kafka and Iceberg are surfaced as one logical view when you query.
• Fresh the moment it lands: no batch windows to wait for.
• No compaction/snapshot catch-up from constant rewrites.
• One logical source: no duplicate tables to reconcile.
• Translate at read time so analytic scans prune effectively, without re-partitioning Kafka.
Keep Your tools. Get Fresher Data.
Query the unified view from any Iceberg-compatible engine or JDBC/ODBC BI tool.
• Bring your own engine: Spark, Trino, Snowflake...
• Works with BI: JDBC/ODBC into dashboards.
• Standard SQL: not a new query language.
• Govern once: shared views and ACLs across tools.
Purpose Built for
Real-time Analytics
Dashboard and Reporting
Keep dashboards and reports aligned with live data. Every Kafka topic is instantly available in Iceberg, so teams can query fresh events without waiting for pipelines to finish.
Machine Learning & AI
Models are never stale. Feature stores and predictions are updated with the most recent data while retaining the full history..
Unified Governance
We carry Kafka’s metadata, schema rules, and access controls through into Iceberg so the data lake stays clean, trustworthy, and auditable.
Data Science Exploration
Queries combine last second activity with years of history in one place. Data scientists can join and analyse streams with accuracy and freshness guaranteed.
What a Unified Kafka + Iceberg View Enables
Structured Table From Raw Kafka
Kafka topics become clean, query-ready tables - translated at read time for fast analytic scans, no repartitioning required.
The Freshest Data
The moment data lands in Kafka, it is queryable in Iceberg. Dashboards, investigations and AI/ML models always stay in step with the stream.
Single Source of Truth
Kafka’s access rules, ACLs and retention windows carry over directly. The same policies apply whether you query a second ago or a year back.
Simplified Operations
No compaction jobs, no snapshot cleanup, no repartition rewrites. Data stays in Kafka, and Iceberg is just a logical view.
Stream Processors Give You Speed, But Not Insights.
OLAP databases provide insights, but never in real time. Streambased combines both: the immediacy of stream processing with the depth of analytics.
Compare
Plug directly into your existing stack.
Integrate directly with Kafka and analytics tools — and move from data to action instantly.
Insights Hub
Stay ahead with perspectives on real-time analytics, streaming architectures, and industry trends.
Sign up and get instant access to the demo
Use Streambased right in your browser, or install it locally in minutes.
Try it out