Penguin Solutions Advances CXL Memory Adoption
Penguin Solutions announced that its SMART Modular CXL NV-CMM E3.S 2T non-volatile memory module passed CXL compliance testing and has ...
In Focus: CXL
| 🖥️ Compute Express Link (CXL) Open Memory Interconnect for AI Infrastructure | |
| Compute Express Link (CXL) is an open, cache-coherent interconnect built on the PCI Express physical layer that enables processors, accelerators, memory devices and storage systems to share memory efficiently. CXL is designed to improve resource utilization and scalability in AI servers, cloud infrastructure and high-performance computing systems. | |
| Why It Matters | AI workloads increasingly demand larger memory pools and faster data sharing between CPUs, GPUs and specialized accelerators. CXL enables memory expansion, pooling and coherent data sharing, helping improve system utilization while reducing memory bottlenecks in next-generation AI infrastructure. |
| Category | Interconnect |
| Key Capabilities | Cache coherency • Memory pooling • Memory expansion • Resource disaggregation • Shared memory • PCIe compatibility • Accelerator connectivity • Rack-scale architecture support |
| Topics Covered | AI Infrastructure • Memory Systems • PCIe • GPU Servers • Data Centers • Rack-Scale Computing • Cloud Infrastructure • High-Performance Computing |
| Key Companies | AMD • Intel • NVIDIA • Astera Labs • Marvell • Broadcom |
| Related Technologies | PCI Express (PCIe) • UALink • NVLink • AI Infrastructure |
| Related Organizations | CXL Consortium • PCI-SIG |
Marvell Expands CXL with CPU and DRAM Interoperability
Marvell announced that its Structera Compute Express Link (CXL) memory-expansion controllers and near-memory accelerators have completed interoperability testing with DDR4 ...
Montage Technology Samples PCIe 6.x/CXL 3.x Retimer Chips
Montage Technology has launched sampling of its latest PCIe® 6.x/CXL® 3.x Retimer, designated M88RT61632. Designed for high-bandwidth applications such as ...
Marvell intros Structera CXL for Enhanced Memory Performance
Marvell introduced its new Structera product line of Compute Express Link (CXL) devices, aimed at resolving memory performance and scaling ...
Samsung builds CXL infrastructure certified by Red Hat
Samsung Electronics has built a Compute Express Link (CXL) infrastructure certified by Red Hat. Elements that configure servers, from CXL-related ...
Astera demos PCIe optical connectivity for expansive GPU clusters
Astera Labs has demonstrated end-to-end PCIe optical connectivity to provide unprecedented reach for larger, disaggregated GPU clusters. The company’s technology ...
Alphawave Semi tapes out multi-protocol I/O connectivity chiplet
Alphawave Semi has announced the successful tape-out of the industry’s first off-the-shelf multi-protocol I/O connectivity chiplet, utilizing TSMC’s 7nm process. ...
Compute Express Link (CXL) 3.1 spec released
The CXL Consortium announced the release of the Compute Express Link (CXL) 3.1 specification with improved fabric manageability to take ...
CXL Memory Expansions Will Transform Data Centers
The emerging CXL standard could be the key to scaling AI platforms and cloud infrastructure effectively. Thad Omura, SVP, Business ...
Tech Update: Breaking through the Memory Wall
The memory wall has been a major challenge in scaling computing performance. This is about to change. Ahmad Danesh, Associate ...
Video: Solving Distributed Computing at Scale
A radical transformation is coming for data center infrastructure as computing systems scale up and scale out. Enfabrica's Accelerated Compute ...
Samsung debuts first 128 GB DRAM with CXL 2.0
Samsung Electronics introduced the first 128-gigabyte (GB) DRAM to support Compute Express Link (CXL) 2.0. The new CXL DRAM supports ...
Astera Labs raises $150M for its CXL platform
Astera Labs, a start-up based in Santa Clara, California, raised $150 million in Series-D funding with a $3.15B valuation for ...
A private dossier for networking and telecoms