The Curious Engineer

All blogs / Databases

Feldera Blog: Incremental View Maintenance and Real-Time SQL

feldera.com/blog

Technical posts on incremental view maintenance, incremental computation, and real-time analytics from the team behind DBSP. About 90 posts and counting.

DatabasesDistributed systems / cloud

’23’24’25’26
Posts
93 from 14 Jul 2023 to 20 Oct 2026
Posting
monthly (12 in the last year)
Kind
company blog

2026 10

  1. Read MoreIntroducing Feldera Scale-Out: Get to Real Time in Half the TimeUse scale-out to spread the one-time read of historical data across as many hosts as needed with a single setting, and get sub-second results sooner
  2. Read MoreAnnouncing our Series A and SeedWe raised $21.5M across a Series A led by Inovia Capital and a Seed led by Costanoa Ventures, to reduce database compute costs by 95% with a mathematical breakthrough.
  3. Read the customer story→
  4. Read MoreAgents Aren’t Coworkers, Embed Them in Your SoftwareAgents do not need more human-like conversation. They need software that makes conversation unnecessary. Give them the right interfaces and they stop acting like noisy copilots and start acting like infrastructure.
  5. Read MoreMaking samply profiles even more usefulFeldera cut one customer's backfill time from 20 hours to 4 by getting serious about profiling. In this post, we share how we made samply profiles significantly more useful by postprocessing their output to add application-level markers — no changes to
  6. Read MoreTurns out we didn’t need that second indexBuilding an incremental compute engine means constantly asking whether the runtime is doing more work than it needs to. This time the answer was yes, and the fix made every pipeline that shares a join source cheaper to run.
  7. Read MoreWhy incremental aggregates are difficult - part 1Many traditional query engines may be able to handle only some kinds of queries, or only some kinds of input updates incrementally, reverting to full recomputation for unsupported operations. Feldera uniformly handles insertions and deletions
  8. Read MoreNobody ever got fired for using a structRust structs are usually the obvious way to represent data. But when you serialize wide SQL tables with hundreds of nullable columns, that "obvious" layout can quietly double your storage cost. Fixing it turns out to be a surprisingly simple trick.
  9. Read MoreCan your incremental view maintenance engine do this?Handle 217 joins, 33 materialized views, 27 aggregations, and 287 linear operators on a single 16-core machine using 15GB RAM at steady state. Here's the proof.
  10. Introducing Feldera Health

2025 27

  1. Feldera in 2025: Building the Future of Incremental Compute
  2. A Visual Profiler for Feldera Pipelines
  3. Constant folding in Calcite
  4. The Dirty Secret of IVM: You Still Need Batch
  5. How Feldera Customers Slash Cloud Spend (10x and beyond)
  6. Stream integration: why databases, bank accounts, and brains all work the same way
  7. The Babel tower of SQL dialects
  8. Batch Analytics at Warp Speed: A User Guide
  9. Toward a real-time medallion architecture
  10. Universal IVM: Incremental View Maintenance Modern Data Stack
  11. Incremental Update 23 at Feldera
  12. Incremental Update 22 at Feldera
  13. Cutting Rust Compile Times from 30 to 2 Minutes
  14. Incremental Update 21 at Feldera
  15. Feldera: three tools for the price of one
  16. Incremental Update 20 at Feldera
  17. The Pain That is Github Actions
  18. Incremental Update 19 at Feldera
  19. Object stores in Feldera pipelines: S3, Azure Blob, GCS
  20. Incremental Update 18 at Feldera
  21. Analyze OpenTelemetry Data in Real Time with SQL
  22. Incremental Update 17 at Feldera
  23. Announcing S3-backed pipelines
  24. Incremental Update 16 at Feldera
  25. Incremental Update 15 at Feldera
  26. Fine-Grained Authorization with Recursive SQL
  27. Incremental Update 14 at Feldera

2024 46

  1. Incremental Update 13 at Feldera
  2. Reflecting on a Year of Innovation and Looking Ahead
  3. Implementing Batch Processes with Feldera
  4. Incremental Update 12 at Feldera
  5. How Feldera Fault Tolerance Works: Technical Details
  6. Correctness at Feldera
  7. Recursive SQL queries in Feldera
  8. Incremental Update 11 at Feldera
  9. Fault tolerance now in preview!
  10. Inspecting Feldera pipelines
  11. Incremental Update 10 at Feldera
  12. Feature Wars: SQL vs DSL for Feature Engineering
  13. Analyze unbounded time-series data with bounded state
  14. Incremental Update 9 at Feldera
  15. Taking the short path to streaming on the GPU with DBSP
  16. Incremental Update 8 at Feldera
  17. LATENESS in streaming programs
  18. Incremental Update 7 at Feldera
  19. Announcing our rebrand, new website, and introducing Fred.
  20. Where was Waldo (when his card was swiped)?
  21. Incremental Update 6 at Feldera
  22. Feldera performance on Nexmark versus Flink
  23. Incremental Update 5 at Feldera
  24. Analyzing query performance in Feldera
  25. Incremental Update 4 at Feldera
  26. Meet the Feldera storage engine
  27. Incremental Update 3 at Feldera
  28. Feldera's new storage engine analyzes more data in less memory
  29. Incremental Update 2 at Feldera
  30. Why We Moved Off RocksDB for Streaming State
  31. Rolling Aggregates for Real-Time Fraud Detection
  32. Optimizing Kafka input performance
  33. Your streaming analytics needs more state than you think
  34. Real-time fraud detection using Feldera and Hopsworks
  35. Let's make streaming analytics boring
  36. Real-time feature engineering with Feldera. Part 2.
  37. What do we mean when we say real-time SQL?
  38. Incremental Computation, a bad case of déjà vu
  39. SecOps solution using CloudTrail, EventBridge and Feldera
  40. Real-time feature engineering is hard
  41. Streaming queries without compromise
  42. DBSP named an ACM SIGMOD Research Highlight
  43. A synchronous streaming model
  44. Feldera input and output connectors
  45. Incremental Database Computations
  46. Indexed Z-sets: JOIN and GROUP BY on changes

2023 10

  1. Database computations on Z-sets
  2. Calcite program representations
  3. How to Continuously Analyze Streaming Data from Redpanda
  4. Desperately Seeking Kafka: consumer seek behaviour
  5. Implementing Z-sets in Java, a Step by Step Guide
  6. Z-sets: representing database changes
  7. DBSP wins Best Research Paper, VLDB 2023
  8. Announcing Feldera, the company
  9. Computing with changes: incremental view maintenance
  10. Welcome To Feldera