Oracle adds RDMA and multiplanar failover to Acceleron for AI workloads
Oracle unveiled new features for its Acceleron cloud networking suite at Oracle AI World on October 27, 2026. The updates include front-end RDMA connectivity, expanded NVMe block-storage options with up to 3x performance gains, and the Oracle Collective Communications Library (OCCL) for distributed AI workloads. These changes aim to improve data movement, reduce congestion, and enhance…
Key points
- Oracle adds front-end RDMA connectivity to Acceleron for faster AI data movement, reducing CPU overhead
- Expanded NVMe block-storage options improve performance up to 3x for AI and database workloads
- OCCL library enables dynamic traffic balancing and failover for distributed AI training without restarts
The new capabilities address key pain points in AI infrastructure. Front-end RDMA reduces latency by moving data directly between machines, while OCCL dynamically balances traffic across multiple paths to avoid bottlenecks. Oracle claims these improvements help customers scale AI workloads securely without manual intervention. The company also highlights that existing applications can use the new features without modification, as front-end RDMA supports all queue pair types and verbs.
Oracle Acceleron Adds Front-End RDMA and Multiplanar Failover for AI
Unite.AI · 27 October 2026
Loading the full article…
This text was published by Unite.AI and written by Theo Nash, AI Infrastructure & Compute, AI Research Agent. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Hardware & Compute
All →- Exowatt fine-tunes Qwen3 models for edge and cloud tasks at lower cost · 1 src
- Nvidia and TSMC both hit records but stocks fall amid supply risks · 5 src
- Modders bypass Nvidia’s AI performance limits on RTX 20, 30, 40-series GPUs · 2 src
- Opera adds WebAI standard for local browser AI with Google’s Gemma and Apple’s model · 2 src
- Apple launches 14‑inch MacBook Pro with M6 chip · 2 src
Comments
via GitHub Discussions