Panmnesia samples first PCIe 6.0/CXL 3.2 fabric switch for Artificial Intelligence infrastructure

Panmnesia has begun sampling a hybrid PCIe 6.0/CXL 3.2 fabric switch that supports port-based routing. The switch targets Artificial Intelligence infrastructure and remains backward compatible with earlier PCIe and CXL generations.

Panmnesia, a pioneer in Artificial Intelligence infrastructure link solutions, announced sample availability of its PCIe 6.0/CXL 3.2 fabric switch, which the company says is the industry first CXL switch silicon to support port-based routing. Panmnesia is distributing switch silicon samples to early access partners as part of the roll out. The announcement positions the device as a new option for system builders and vendors working on next generation interconnects for data center workloads.

The switch is described as a hybrid solution that implements both PCIe Gen 6 and CXL 3.2 protocols on a single chip while maintaining full backward compatibility with all previous PCIe and CXL generations. According to Panmnesia, the implementation follows the CXL 3.2 and PCIe 6.0 specifications strictly, which the company says ensures standards compliance. The single-chip approach aims to simplify platform design by handling multiple interconnect protocols in one device.

In addition to supporting standard routing, the fabric switch can operate in port-based routing mode and hierarchy-based routing mode, giving designers flexibility in how devices are interconnected. Panmnesia emphasizes complete interoperability across all subprotocols as a key feature, noting that the silicon is fully implemented in accordance with the relevant specifications. The company is moving samples to early partners to validate real-world interoperability and integration with existing server and accelerator ecosystems.

58

Impact Score

NVIDIA sweeps MLPerf training v5.1 for artificial intelligence

NVIDIA swept all seven tests in MLPerf Training v5.1, posting the fastest training times across large language models, image generation, recommender systems, computer vision and graph neural networks. The company was the only platform to submit results on every test, highlighting its GPUs and CUDA software stack.

NVIDIA sweeps MLPerf Training v5.1 benchmarks

NVIDIA posted the fastest times across all seven MLPerf Training v5.1 tests, demonstrating new gains in Artificial Intelligence training from GPUs to networking and software. The company submitted results on every benchmark and introduced Blackwell Ultra and NVFP4-backed optimizations.

When Artificial Intelligence meets biology

Microsoft researchers disclosed a confidential study that explored how open-source Artificial Intelligence tools could bypass biosecurity checks and said their work helped produce fixes now influencing global standards.

Contact Us

Got questions? Use the form to contact us.

Contact Form

Clicking next sends a verification code to your email. After verifying, you can enter your message.