Compute-in-memory (CiM) is one promising solution to address the memory bottleneck existing in traditional computing architectures. However, the tradeoff between energy efficiency and computing precision plagues most CiM implementations, and the low precision imposes a major limitation on CiM’s ability to support practical computational workloads. In this article, a static random access memory (SRAM)-based analog CiM macro is presented with the Intel 22FFL process. By introducing a 1-to-2 ratioed capacitor ladder (C-2C)-based charge domain computing scheme, the proposed CiM prototype chip demonstrates a maximum of 2k multiply-accumulate (MAC) operations in one clock cycle and achieves 32.2-TOPS/W peak power efficiency with 8-bit precision in both input activation and weight while ensuring accurate on-chip matrix–vector multiplications (MVMs) with a computation error less than 0.5%. A 4.0-TOPS/mm $^{2}$ peak area efficiency is attained by adopting a local weight multiplexing scheme with a 9T SRAM cell, which improves the memory density and reduces the need to refresh the weight stored in the SRAM array. A variety of analog impairment factors, including parasitics, mismatch, and noise, were analyzed to guarantee a sufficiently high multibit linearity. The proposed passive analog computing mechanism ensures the computation accuracy over process–voltage–temperature (PVT), with the measured MVM error deviation of less than 1% over PVT variations.
22 October 2021 in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
Accurate characterization of radiation-induced soft errors is a critical step toward understanding the impact of these glitches on circuit and system reliability. With process scaling, there has been exponential increase in number of transistors that can be packed on a die which, in turn, results in higher sensitive node count and persistent soft error susceptibilities. In this work, a novel circuit technique employing higher sensitivity toward soft errors is proposed. The circuit makes use of current-starved gates with bias knobs to fine-tune both measurement resolution and strike sensitivity enabling accelerated and efficient induction of errors in a limited-time irradiation test environment. The back-sampling chain (BSC) circuit can measure individual radiation-induced transient pulse with as low amplitude as 0.3 x VDD while maintaining a high measurement resolution for pulsewidth characterization. The bias knobs allowing tuning of sensitivity and resolution enable, for the first time, a strike pulse waveform reconstruction methodology that can be used to calibrate current pulse models for assessing soft error rate (SER) sensitivity of standard logic gates.
24 October 2025 in IEEE Journal of Solid-State Circuits
The increasing adoption of chiplet-based architectures highlights the compelling paradigm of disaggregating monolithic systems into interconnected chiplets, offering numerous advantages for heterogeneous integration. This disaggregation is essential as chips approach the reticle limit, improving yield and design flexibility by combining diverse functionalities. Next-generation 2.5-D systems require enhanced interoperability across heterogeneous multi-foundry nodes, higher memory, I/O, and scalable interconnects for varied chiplet types driven by artificial intelligence (AI) and data-intensive workloads. We introduce a unique 2.5-D packaging approach integrating heterogeneous multi-foundry active chiplets over a large passive base in silicon (PBS) chiplet. Our system in package (SIP) houses up to 20 (${\sim } 4\,{\times }\,4$ mm) chiplets, demonstrating significant and functional heterogeneous integration and scalability beyond reticle limits, addressing integration complexities. This platform enables the creation of highly customized compute systems tailored to specific application demands. Furthermore, the high bandwidth and energy efficiency facilitate complex data processing and acceleration. To tackle memory bottlenecks in AI, future systems should enable workload-tailored assembly for optimal heterogeneous resource allocation. Our SIP achieves 20-Tb/s aggregate bandwidth in parallel computing across heterogeneous units with 0.8-pJ/bit die-to-die (D2D) efficiency, showcasing potential for high-performance, power-conscious heterogeneous disaggregated systems.