Fujitsu has showcased the first silicon wafers and an engineering sample of its ‘Monaka’ CPU during MWC, developed in partnership with networking equipment maker 1FINITY. Scheduled for release in 2027, the initial Fujitsu Monaka CPU utilizes the Armv9-A architecture and a 3D chiplet layout that combines a core die with separate SRAM and I/O dies for improved modularity and scalability. The company positions the system on chip as a next generation platform for Artificial Intelligence inference, simulation, and large scale data processing, with the intention to sell complete systems to external customers.
A single chip features 144 cores, and two-socket configurations can scale up to 288 cores per node, targeting dense compute environments. The platform supports 12-channel DDR5, PCIe 6.0 with CXL 3.0, and Arm SVE2 for Artificial Intelligence and high performance computing workloads, aligning the design with modern memory bandwidth and interconnect standards. Fujitsu has chosen TSMC to manufacture this chip using the 2 nm node, paired with Broadcom’s 3.5D eXtreme Dimension System-in-Package (XDSiP) packaging architecture, which focuses on advanced integration of multiple dies and memory around the processor. This packaging allows Monaka to become a 144-core design featuring four 36-core chiplets, and these chiplets are stacked face-to-face with SRAM tiles through hybrid copper bonding, utilizing TSMC’s N5 process for the cache layer.
In early sample packaging images, the silicon complex shows a large central I/O die, HBM memory arranged around the CPU, and the new packaging technology implemented on the package. Reportedly, this CPU has already reached a working version, with Broadcom shipping the CPU to Fujitsu in late February this year, enabling the start of bring up and validation work. After initial testing and early performance validation, Fujitsu plans to ship these processors to customers around summer, with mass shipping to commence in 2027 as production ramps. Interest is expected from users who previously adopted Fujitsu’s A64FX, which powered the Fugaku supercomputer that was the most powerful system back in 2020, achieving 415.53 PetaFLOPS of FP64 and an HPL-AI score of 1.421 ExaFLOPS using lower FP16 precision, and the new Monaka CPU is expected to enable much greater speeds with efficiency improvements.
