Latest Marvell Blog Articles
-
September 15, 2026
Breaking Through the KV Cache Memory Wall with DPU-powered Network Storage
By Chander Chadha, Director of Product Marketing, Storage Products, Marvell
As AI models grow larger and inference workloads scale, larger models, longer context windows and growing KV caches are driving demands on memory resources. Traditional CPU-centric networked JBOF (Just Bunch of Flash) can’t keep pace with the throughput, latency, and efficiency requirements of these modern AI clusters.
DPU (data processing unit) -based storage is a compelling alternative to traditional storage, acting as the broker between the network and SSD for remote storage. By offloading storage, networking and security processing from the host CPU onto a dedicated DPU, AI infrastructure can move data closer to compute, reduce latency, and free up valuable CPU cycles for AI workloads.
To address this need, Marvell offers the OCTEON DPU family, purpose-built for hyperscale cloud workloads and data center applications, extending its use specifically into network storage acceleration for AI environments.
-
September 10, 2026
Marvell Demonstrates Scalable CXL Memory Infrastructure with Intel
By Khurram Malik, Associate Vice President, Custom Cloud Solutions, Marvell
The rapid growth of AI is creating unprecedented demand for memory capacity. As models get larger and workloads become more data-intensive, traditional server architectures are increasingly challenged to scale memory efficiently.
At Flash Memory Summit (FMS), Marvell demonstrated how CXL can enable a more flexible and scalable approach to memory infrastructure, showcasing an end-to-end architecture at the Intel® booth.
The demonstration combined an Intel platform with three Marvell® technologies: Structera® X CXL memory expander, Structera® S CXL switch and Alaska® P PCIe retimer. Together, they demonstrated how memory expansion, CXL switching and high-speed PCIe connectivity can work together within a real server platform.
-
September 09, 2026
Marvell Structera X Extends CXL Interoperability to NVIDIA Vera
By Arifur Rahman, Director of Product Marketing, Custom Cloud Solutions, Marvell
Memory has become the defining constraint of modern AI infrastructure. Large language models, in-memory databases, and deep learning recommendation models all share the same bottleneck: there is never enough DRAM. Compute Express Link (CXL) was designed to break that bottleneck but a CXL memory expander is only as valuable as the breadth of platforms it can run on.
That is why ecosystem enablement is a core pillar of Marvell's CXL strategy. Today we are marking a new milestone: successful interoperability of the Marvell® Structera™ X CXL memory-expansion controller with the NVIDIA Vera CPU platform.
-
September 08, 2026
Engineered to Collaborate: Robot Racing as a Microcosm of Marvell
By Vienna Alexander, Marketing Content Professional, Marvell
“You have exactly one hour. Good luck! Go!!” the announcer shouted into the microphone. At the Marvell headquarters in Santa Clara, California, the cafeteria was transformed into an engineering workshop. Every table was filled with a box of mechanical parts containing everything needed to build a robot fit for a race around campus.
Around 170 Marvell employees teamed up in groups of eight to ten to bring their visions to life. The Robot Racing event, a Marvell tradition, sparks creativity and teamwork among colleagues, embodying what Marvell is all about.
“It’s a great opportunity for us to get together, use our brains outside of regular meetings, have a team building event, laugh together and have fun,” said Principal Engineer Anis Weldon. “And if we win? That’s good! And even if we lose, that’s okay. We get the experience.”
-
September 01, 2026
Program Spotlight: Marvell New Graduates Accelerate Professional/Technical Skills and Build Community
By Vienna Alexander, Marketing Content Professional, Marvell
Entering the workforce as a new graduate is a big milestone. Marvell aims to set up new grads for success through supportive programming designed to fast-track early career growth and foster community among cohorts.
Approximately 200 new graduate hires across the United States, Canada, and India recently completed the inaugural New Graduate Accelerator Program at Marvell. Through technical learning, professional development and community-building experiences—including connections with senior leaders and engineers across the company—new grads built the technical expertise, skills and relationships needed to navigate the first year at Marvell. Along the way, we also gained a deeper understanding of the company’s business, technology and culture.
Recent Posts
- Breaking Through the KV Cache Memory Wall with DPU-powered Network Storage
- Marvell Demonstrates Scalable CXL Memory Infrastructure with Intel
- Marvell Structera X Extends CXL Interoperability to NVIDIA Vera
- Engineered to Collaborate: Robot Racing as a Microcosm of Marvell
- Program Spotlight: Marvell New Graduates Accelerate Professional/Technical Skills and Build Community
