InfoQ Homepage Streaming Content on InfoQ
-
Airbnb Cuts Recommendation Staleness from 2 Days to Seconds with Chronon
Airbnb extended its Chronon feature platform with Push Mode and Near Real-Time Model Transform to update sequence-based recommendation embeddings within seconds. The architecture reduced embedding staleness from nearly two days to under a minute and improved offline NDCG by 1.67%, while online testing increased uncancelled bookings by approximately one-third of a percent.
-
Pinterest Uses Event Time Watermarks to Finalize Petabyte Scale Data Partitions
Pinterest has introduced event time-based partition finalization for its database ingestion framework. The mechanism uses Apache Flink to track event time and Apache Iceberg metadata to publish a non-regressing watermark, helping downstream jobs determine when continuously ingested CDC data is sufficiently complete for processing.
-
Cloudflare K2 Builds Event Streams on R2 Object Storage, at a Second of Produce Latency
Cloudflare has launched K2 in public beta, a serverless event streaming service implementing a durable log on R2 object storage. Cloudflare states produce latency of about one second at p99; a tester reported end-to-end p99 of 7.5 seconds, which a commenter answering for the team said was higher than expected. Consume charges match produce charges per GB.
-
OpenAI Details GPT-Live’s Architecture for Continuous Stateful Voice Interaction
OpenAI recently published an engineering account of GPT-Live. It described how they designed the system to maintain continuous voice interaction while separating latency-sensitive media processing from broader application work. The live path contains the media pipeline and inference loop, while delegation, tool use, persistence, and other application logic run behind an asynchronous RPC boundary.
-
Nuxt 4.5: Experimental SSR Streaming, Vite 8 and an Rsbuild-Powered Rspack Builder
Nuxt has released version 4.5, featuring updates such as a switch to Vite 8, a new Rspack 2 builder, and experimental SSR streaming. This streaming enhances Time to First Byte by flushing the HTML shell instantly. The release also includes a stable error code system and new composables, alongside important upgrade instructions for developers moving from earlier versions.
-
JioHotstar Explains the Distributed Engineering behind Personalized Ad Requests at Streaming Scale
JioHotstar explains the distributed architecture behind its real-time ad request workflow, covering ad decisioning, waterfall tiering, pacing algorithms, latency optimization, and service coordination required to select and deliver personalized advertisements during streaming playback at scale.
-
Inside Atlassian’s Forge Billing Architecture for Distributed Usage Tracking at Scale
Atlassian details the Forge billing platform built for usage-based pricing across its cloud ecosystem. It processes large-scale usage events with correct attribution, deduplication, and aggregation using a streaming pipeline, idempotent processing, and layered storage to enable accurate billing, near real-time visibility, and reliable reconciliation across distributed services.
-
Confluent Moves Schema IDs to Kafka Headers to Simplify Schema Governance
Confluent introduces a new approach in Apache Kafka that moves schema IDs from message payloads to record headers, aiming to simplify schema governance and evolution. The update integrates with Schema Registry, improves compatibility across serialization formats, and reduces coupling between data and metadata in event-driven architectures.
-
QCon London 2026: Introducing Tansu.io — Rethinking Kafka for Lean Operations
Peter Morgan introduced Tansu at QCon London, an open-source, Kafka-compatible, stateless, leaderless broker that scales to zero, with pluggable storage (S3, SQLite, Postgres), broker-side schema validation, and direct writes to Iceberg and Delta Lake. Written in Rust, it uses 20MB of RAM and starts in 10 milliseconds.
-
From On-Demand to Live : Netflix Streaming to 100 Million Devices in under 1 Minute
Netflix’s global live streaming platform powers millions of viewers with cloud-based ingest, custom live origin, Open Connect delivery, and real-time recommendations. This article explores the architecture, low-latency pipelines, adaptive bitrate streaming, and operational monitoring that ensure reliable, scalable, and synchronized live event experiences worldwide.
-
Airbnb’s Mussel V2: Next-Gen Key Value Storage to Unify Streaming and Bulk Ingestion
Airbnb’s engineering team re-architected its internal key-value storage system, Mussel, to unify streaming and bulk ingestion while simplifying operations, achieving over 100,000 writes per second and sub-25ms read latencies on 100-terabyte tables, while leveraging Kubernetes, Kafka, and a NewSQL backend to improve scalability, reliability, and operational efficiency across its internal services.
-
AWS Lambda Response Streaming Increases Payload Limit to 200 MB
AWS has revolutionized Lambda with an increased response streaming payload limit from 20 MB to 200 MB. This enhancement allows developers to stream larger data sets, improving Time to First Byte performance. By simplifying response handling and eliminating complex workarounds, AWS empowers developers to deliver rich content seamlessly, transforming serverless applications.
-
AWS Lambda Gains Native Avro and Protobuf Support for Kafka Events with Schema Registry Integration
Lambda now natively supports Apache Avro and Protobuf events, streamlining Kafka event processing - an enhancement that eliminates the need for custom deserialization, automates schema validation and filtering, and optimizes costs through efficient event handling. Integration with AWS Glue and Confluent registries simplifies development, allowing cleaner data consumption and enhanced scalability.
-
Confluent Cloud Streamlines Real-Time AI with New Flink Capabilities
Confluent Cloud now simplifies real-time AI development with Flink Native Inference, Flink search, and built-in ML functions—streamlining workflows and reducing costs. This unified Flink SQL solution enhances security and efficiency by keeping data secure within the platform, empowering developers to deploy AI applications quickly and effectively.
-
Amazon GameLift Streams Launches for High-Fidelity, Browser-Based Game Streaming
Introducing Amazon GameLift Streams: a revolutionary service that empowers developers to stream games directly to WebRTC-enabled browsers at 1080p and 60fps. Simplify deployment, reach a broad audience, and enhance player engagement. No infrastructure hassles or code overhauls needed—just instant access to high-fidelity gaming. Transform your titles today!