Broadcom Aims to Solve the Distributed AI Networking Challenge

---

https://www.electronics-lab.com/broadcom-aims-to-solve-the-distributed-ai-networking-challenge/

The new Jericho4 router uses 3.2T HyperPorts and deep buffering to enable lossless RoCE performance, connecting massive XPU clusters across facilities separated by over 100 km.

Memory Fabric System Tackles AI Inference’s Memory Bandwidth Challenges

---

https://www.electronics-lab.com/memory-fabric-system-tackles-ai-inferences-memory-bandwidth-challenges/

Enfabrica's new EMFASYS system integrates CXL-based DDR5 memory with RDMA Ethernet to reduce inference costs for large-scale AI workloads through an elastic memory fabric.

Veyron V2 RISC-V Platform is Designed for AI and Data Center Workloads

---

https://www.electronics-lab.com/veyron-v2-risc-v-platform-is-designed-for-ai-and-data-center-workloads/

Ventana Micro Systems Inc., a well-known designer of RISC-V platforms, has recently announced the release of Veyron V2 RISC-V Platform, an advanced accelerated compute platform designed for AI, data centers, and emerging workloads. The main function of this SoC is that it integrates third-party IPs and domain-specific accelerators, making it highly adaptable across industries like […]