InfoQ - Streaming https://www.infoq.com InfoQ Streaming feed Presentation: From S3 to GPU in One Copy: Rethinking Data Loading for ML Training https://www.infoq.com/presentations/vortex-columnar-file-format-gpu-streaming/?utm_campaign=infoq_content&utm_source=infoq&utm_medium=feed&utm_term=Streaming <img src="https://res.infoq.com/presentations/vortex-columnar-file-format-gpu-streaming/en/mediumimage/onur-satici-medium-1787813397611.jpeg"/><p>Onur Satici explains how Vortex, an open-source columnar file format under the Linux Foundation, revolutionizes high-throughput data loading. He details how cascading lightweight encodings, layout-based segment pruning, and zero-copy memory pipelines eliminate CPU/NVMe bottlenecks to stream S3 data straight to GPUs at speeds up to 60 Gbps without requiring upfront data reprocessing.</p> <i>By Onur Satici</i> GPU Data Lake Performance Data Pipelines Rust Architecture CUDA Transcripts Streaming QCon London 2026 Columnar Databases S3 Machine Learning Architecture & Design Development AI, ML & Data Engineering presentation Fri, 04 Sep 2026 11:00:00 GMT https://www.infoq.com/presentations/vortex-columnar-file-format-gpu-streaming/?utm_campaign=infoq_content&utm_source=infoq&utm_medium=feed&utm_term=Streaming Onur Satici 2026-09-04T11:00:00Z /presentations/vortex-columnar-file-format-gpu-streaming/en OpenAI Details GPT-Live’s Architecture for Continuous Stateful Voice Interaction https://www.infoq.com/news/2026/09/openai-gpt-live/?utm_campaign=infoq_content&utm_source=infoq&utm_medium=feed&utm_term=Streaming <img src="https://res.infoq.com/news/2026/09/openai-gpt-live/en/headerimage/OpenAI-gpt-live-system-architecture-header-1788176199388.jpeg"/><p>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.</p> <i>By Eran Stiller</i> RPC OpenAI Streaming Voice-enabled UI WebRTC ChatGPT Architecture & Design news Wed, 02 Sep 2026 12:20:00 GMT https://www.infoq.com/news/2026/09/openai-gpt-live/?utm_campaign=infoq_content&utm_source=infoq&utm_medium=feed&utm_term=Streaming Eran Stiller 2026-09-02T12:20:00Z /news/2026/09/openai-gpt-live/en