Marvell and Penguin Launch Photonics Memory Appliance
<p>One of the biggest impediments to progress in AI these days is memory. Customers just don’t have memory that’s big enough and fast enough to handle their multi-rack AI inference workloads, which limits the speed and volume of AI token generation. To address that need, today at SC26, Marvell and Penguin Computing launched a Photonics […]</p>
<p>The post <a href="https://www.hpcwire.com/2026/11/16/marvell-and-penguin-launch-photonics-memory-appliance/">Marvell and Penguin Launch Photonics Memory Appliance</a> appeared first on <a href="https://www.hpcwire.com">HPCwire</a>.</p>
Read original article ↗
Related Articles
Marvell Scales Connectivity to Address AI Infrastructure’s Next Bottleneck at SC26
As AI Workloads Outgrow Single-rack Memory and Bandwidth, Marvell Demonstrates Interconnect and Memory Technologies Engi
Nokia eyes AI optical interconnect market, potentially challenging Lumentum and Coherent
The AI data center optical interconnect market may soon see another heavyweight contender. Roughly a week after Marvell
HBM smuggling routes into China; Nvidia H200s held in Hong Kong; Micron pauses Taiwan fab work; Samsung expands HBM base
Marvell and Lumentum Announce Combination
Marvell Technology, Inc. and Lumentum today announced that they have entered into a definitive agreement for Marvell to