Penguin Solutions Unveils Industry’s First Photonics-Based HBM-Class Memory Appliance to Break the GPU Memory Wall for Agentic AI
Business Wire•16/11/2026•09:00 ET
8
Key Highlights
- ➤Penguin unveils MemoryAI Photonic Appliance for agentic AI memory constraints
- ➤Appliance targets 57.6 Tbps of total bandwidth
- ➤Supports up to 16 TB DDR5 and 576 GB HBM3E per appliance
- ➤Sampling to select customers planned for the second half of 2027
Expert Statements
Andy Mills, VP, Advanced Product Development, Penguin Solutions
“Agentic AI continues to put stress on HBM memory and the GPU, and the MemoryAI™ Photonic Appliance is designed to relieve that stress. Photonics is the step-function change that makes this possible. Instead of hopping out to a CPU over a slow network, a GPU will have a direct way to access memory at scale, and CPUs and GPUs will be able to transparently share terabyte-class memory as though it were directly attached.”
Preet Virk, Senior vice president and general manager, Photonic Fabric Business Unit, Marvell
“For AI inference to scale, we need to deliver tokens with greater economic efficiency. By integrating Marvell Photonic Fabric technology into Penguin Solutions' MemoryAI Photonic Appliance, we are creating a path to high-bandwidth, low-latency and high-capacity memory sharing across multiple racks. This approach is designed to improve accelerator utilization and increase the quality of tokens and throughput.”
Penguin Solutions Unveils Industry’s First Photonics-Based HBM-Class Memory Appliance to Break the GPU Memory Wall for Agentic AI
MemoryAI™ Photonic Appliance, featuring Marvell(®) Photonic Fabric™ technology, is designed to enable the transparent sharing of disaggregated high-bandwidth memory across multiple CPU and GPU-based servers
Get started
Create a free account to read the full story.
