How advanced NoC architectures and coherent subsystem IP can address the industry's next-gen scalability, safety, and ...
First-silicon success falls; engineering capacity; minimum clock period; optimizing PyTorch; counterfeit electronics.
Increasingly complex chip designs require more test data than those developed at older nodes and on single planar dies. The challenge is moving all of that data through the system without slowing down ...
Next-gen LPDDR6 and LPDDR5x memory controller and PHY IP is purpose-built to solve the memory wall at every tier of the AI inference hierarchy, and enables cloud, edge, and physical AI inference ...
More capable robots require systems that realize capabilities through appropriate paradigms, compose them across time, distribute them across heterogeneous compute, and govern them under explicit ...
A scalable LPDDR-based memory platform optimized for edge AI inferencing.
SW co-design in the AI era; scaling thermal analysis; 800VDC in AI data centers; future AI compute pushing to heterogeneous clusters; localized edge intelligence in edge and cloud AI; AI and voice ...
Cost per token is causing EDA design budgets to balloon; what comes next isn’t entirely clear yet. Discussions have shifted ...
Data centers are notorious energy consumers. On the one hand, ongoing efforts attempt to reduce electronics power to keep overall power from growing. On the other hand, data centers are provided with ...
Even though the benefits are accepted, a combined hardware/software development flow is hampered by a lot of challenges.
One toolchain means one team. The DSP specialists, the ML engineers, and the firmware developers work in the same environment ...
Build configurable PCIe subsystems spanning local expansion, shared resource domains, and longer-reach topologies.
Some results have been hidden because they may be inaccessible to you
Show inaccessible results