Semiconductors

There's a rising trend of hyperscalers demanding customized silicon for their workloads. What's driving the rush for custom silicon in AI infrastructure? Raghib Hussain, President, Products and Technologies from Marvell...

The latest AI data center projects are insanely large. Can multiple gigawatts for power really be created in the timeline necessary to launch these massive data centers? Loi Nguyen, EVP...

Many types of connectivity now exist within the data center for both copper and fiber. Which ones prevail for future builds? Nick Kucharewski, SVP and GM, Network Switching from Marvell...

In response to requests from hyper scale customers, Marvell pioneering a new class of "coherent-lite" connectivity. Achyut Shah, SVP and GM, Connectivity from Marvell explains: - Coherent-lite bridges the gap...

Google Cloud announced the general availability of its sixth-generation Tensor Processing Unit (TPU), Trillium, which powers the company’s advanced AI Hypercomputer architecture. Designed to tackle the demands of multimodal AI...

Spirent Communications is collaborating with Siemens Digital Industries Software to deliver a pre-silicon verification solution for Ethernet chipset design. The integration combines Spirent TestCenter Virtual with Siemens’ Virtual Ethernet test...

Marvell released a 1.6 Tbps transimpedance amplifier (TIA) and laser driver chipset, enabling 800 Gbps and 1.6 Tbps linear-drive pluggable optics (LPO) for short-reach compute fabric interconnects. The new chipset...

Marvell unveiled the Aquila DSP, the first coherent-lite DSP optimized for 1.6 Tbps O-band optical interconnects. Designed for data center campus connectivity spanning 2 to 20 km, the Aquila DSP...

Marvell introduced a custom high-bandwidth memory (HBM) compute architecture designed to optimize XPUs for cloud AI accelerators. This innovation delivers up to 25% more compute capability and 33% higher memory...

Broadcom has announced its 3.5D eXtreme Dimension System in Package (XDSiP) platform, a new technology designed for next-generation AI accelerators (XPUs). The platform integrates over 6,000 mm² of silicon and...

At AWS re:Invent, Amazon Web Services (AWS) announced the general availability of Trainium2-powered EC2 Trn2 instances and introduced the Trn2 UltraServers, designed for high-performance AI model training and inference. These...

Comcast has transitioned its 5G wireless core network from on-premises infrastructure to Amazon Web Services (AWS). The migration enables Xfinity Mobile and Comcast Business Mobile to deliver advanced 5G connectivity...

Marvell introduced the Ara platform, the first 3nm PAM4 DSP supporting 1.6 Tbps interconnects with 200 Gbps electrical and optical interfaces. Ara cuts optical module power consumption by more than...