Breaking the Memory Wall: How d-Matrix Is Redefining AI Inference with Chiplets
By Maurizio Di Paolo Emilio, embedded.com | April 23, 2025
As AI workloads push the limits of performance, power efficiency, and memory bandwidth, chiplets are rapidly emerging as the architectural solution of choice. In this interview, Sree Ganesan, Vice President of Product at d-Matrix, dives deep into how its pioneering chiplet-based platform is revolutionizing AI inference. From solving the memory wall with Digital In-Memory Computing (DIMC) to enabling seamless multi-chiplet communication via custom interconnects, d-Matrix reveals how its innovations are unlocking 10x faster token generation, 3x better energy efficiency, and a scalable roadmap for generative AI.
To read the full article, click here
Related Chiplet
- Bridglets
- Automotive AI Accelerator
- Direct Chiplet Interface
- HBM3e Advanced-packaging chiplet for all workloads
- UCIe AP based 8-bit 170-Gsps Chiplet Transceiver
Related News
- Alphawave Semi Tapes Out Industry-First, Multi-Protocol I/O Connectivity Chiplet for High-Performance Compute and AI Infrastructure
- YorChip announces Low Latency 200G Chiplet for edge AI
- Tenstorrent Licenses Baya Systems’ Fabric into next-generation AI and Compute Chiplet Solutions
- DreamBig closes $75M Series B Funding Round, Co-led by Samsung Catalyst Fund and Sutardja Family to Enable AI Inference and Training Solutions to the Masses
Latest News
- Development Flows For Chiplets
- Die-to-die Interconnect Standards In Flux
- Secafy Licenses Menta's eFPGA IP to Power Chiplet-Based Secure Semiconductor Designs
- EV Group Hybrid Bonding, Maskless Lithography and Layer Transfer Solutions for Heterogeneous Integration to be Highlighted at ECTC 2025
- Avicena Announces Series B Funding Round Led by Tiger Global with Participation from SK hynix