Advertisement


Home Blog Page 13
Huawei UB-mesh
The third and final machine learning presentation before the afternoon break comes from Huawei. Unlike many of the other ML vendors who are here...
Card Level Scaleup: 16 Chiplet Hierarchical All-to-All
d-Matrix Corsair is an in-memory computing platform for AI Inference and the company went into its future plans at Hot Chips 2025
Why Is Memory Important
Marvell showed its massively dense SRAM, its custom HBM solution, and how it is adding Arm cores to CXL memory controllers at Hot Chips 2025
Fabric8Labs ECAM Enabled Thermal Solutions _Page_11
Fabric8Labs ECAM uses the electrical current from an OLED display to manufacture copper liquid cooling 3D coldplates with pixel precision
NVIDIA Co Packaged Optics With Silicon Photonics At Hot Chips 2025 _Page_14
At Hot Chips 2025, NVIDIA showed its co-packaged optics solutions and its next-generation Spectrum-XGS for scale-across long-distance AI clusters
Lightmatter Passage M1000 At Hot Chips 2025 Passage M1000 Components
The Lightmatter Passage M1000 is an optical interposer that will allow for massive chips with 114Tbps of bandwidth
Ayar Labs UCIe Optical IO Retimer Hot Chips 2025 _Page_06
Ayar Labs has a UCIe optical I/O retimer that it is showing off at Hot Chips 2025. The basic idea is to make a...
Celestial AI Photonic Fabric Module Hot Chips 2025 _Page_08
Celestial AI has a new Photonic Fabric Module that frees chip designers to place high-bandwidth optical interconnects even in the center of packages
Broadcom Tomahawk Ultra Hot Chips 2025 _Page_04
Broadcom is talking about the Tomahawk Ultra, its 51.2T at 77 billion packets per second switch for AI and HPC networks
NVIDIA ConnectX 8 SuperNIC Port At SC24 1
At Hot Chips 2025, NVIDIA went into the ConnectX-8 SuperNIC and some of the features that make it different from others on the market