Additive Manufacturing (AM), also known as 3D printing, offers several advantages, including on-site production, enhanced throughput, and efficient use of raw materials. However, the rise in its usage has also led to an increase in potential threats that aim to disrupt the printing process. These attacks can subtly alter the design (CAD or STL) files or machine instructions (g-code), which can cause significant economic and reputational harm to the victim company. Current detection techniques, based on acoustic, magnetic, and accelerationbased side-channel analysis, have proven to be ineffective. Although power side-channel analysis is more effective than other means, it is expensive and not scalable. This paper proposes a novel detection method, SCREAM, that assumes the user has access to a trusted STL source and an untrusted g-code. SCREAM leverages the pulse trains sent to the motors to reconstruct the executing g-code. To ensure the safe and accurate execution of g-code, a three-level comparison is performed between recovered and untrusted g-code, as well as trusted STL ensuring successful detection of any anomalies present in the executing g-code. Our testing has shown that this method can detect a range of existing attacks on AM, including malicious firmware manipulation, FLAW3D, and Needle in a Haystack.

The semiconductor industry is steadily shifting toward disaggregated system design to overcome the scalability (yield and cost) limitations of monolithic integration. 2.5D integration offers a compelling pathway for realizing such systems, enabling the assembly of heterogeneous chiplets—compute, memory, analog & I/O, with dense interconnects and high bandwidth. Our previous work showcased experimental results from a multi-foundry chiplet design over a large passive silicon base. There were 20 chip slots (CS) on an interposer that can be configured with compute die (CD) or memory die (MD) chiplets. Building on this foundation, this paper introduces a methodology for system technology co-optimization (STCO) across several vectors. These include varying the number of chip slots, configuring slots with different numbers of MDs and CDs, sweeping the inter-die bandwidth, choosing between different technology nodes for the performance limiting MDs. This work demonstrates how the design and technology choices affect system performance, power, and cost across different workloads, empowering designers to select optimal configurations for their specific needs.

Deep neural network (DNN) inference tasks have become ubiquitous workloads on mobile SoCs and demand energy-efficient hardware accelerators. Mobile DNN accelerators are heavily area-constrained, with only minimal on-chip SRAM, which results in heavy use of inefficient off-chip DRAM. With diminishing returns from conventional silicon technology scaling, emerging memory technologies that offer better area density than SRAM can boost accelerator efficiency by minimizing costly off-chip DRAM accesses. This paper presents a detailed design space exploration (DSE) of technology-system co-design for systolic-array accelerators. We focus on practical/mature on-chip memory technologies, including SRAM, eDRAM, MRAM, and 3D vertical RRAM (VRRAM). The DSE employs state-of-the-art optimizations (e.g., model compression and optimized buffer scheduling), and evaluates results on important models including ResNet-50, MobileNet, and Faster-RCNN. Compared to an SRAM/DRAM baseline, MRAM-based accelerators show up to 4.68× energy benefits (57% area overhead), while a 3D VRRAM-based design achieves 2.22× energy benefits (33% area reduction).

High-density 3D techniques (such as wafer bonding and monolithic-3D) show tremendous promise in reducing interconnect lengths and relieving 2D congestion. We propose an enhanced 3D implementation methodology and use it to design an Arm Cortex-A microprocessor in 3D. The methodology is fully integrated and tested using commercial EDA tools and incorporates all physical IP needed to implement modern microprocessors. The resulting 3D implementation consists of two parts, 1) a multi-tier co-placement approach for enhanced placement quality, 2) integration of 3D SRAMs for improved microprocessor PPA. Compared to the 2D baseline, our implementations show an overall area reduction of 8.5% and can either achieve an 18% peak frequency uplift at iso-power or a 41% power reduction at near iso-performance (-3% frequency).

Physical unclonable functions (PUFs) are primitives that generate high-entropy, tamper resistant bits for use in secure systems. For applications such as cryptographic key generation, the PUF response bits must be highly reliable, consistent across multiple evaluations under voltage and temperature variations. Conventionally, error correcting codes (ECC) have been used to improve response reliability, but these techniques have significant area, power, and delay overheads and are vulnerable to information leakage. In this work, we present a highly-reliable, PUF-based, cryptographic key generator that uses no ECC, but instead uses built-in self-test to determine which PUF bits are reliable and only uses those bits for key generation. We implemented a prototype of the key generator in a 65nm bulk CMOS testchip. The key generator generates 1213 bits in an area of2with a measured bit error rate of*10-9in both the nominal and worst case corners (100k measurements each). This is equivalent to a 128-bit key failure rate of-6. The system can generate a 128-bit key in 1.15μs. Finally, we present a realization of a “strong”-PUF that uses 128 of these highly reliable bits in conjunction with an Advanced Encryption Standard (AES) cryptographic primitive and has a response time of 40ns and is realized in an area of 84k/μm2.

Physically Unclonable Functions (PUFs) are structures with many applications, including device authentication, identification, and cryptographic key generation. In this paper we propose a new PUF, called SCAN-PUF, based on scan-chain power-up states. We argue that scan chains have multiple characteristics that make them uniquely suited as a low-cost PUF. We present results from test chips fabricated in a 65nm bulk CMOS process in support of these claims. While approximately 20% of the total population of scan elements are unreliable across temperature variations, we find that simple unanimous selection schemes can result in mean error rates of less than 0.1% for the selected populations across all measurements collected.

The embedded memory hierarchy of microprocessors and systems-on-a-chip plays a critical role in the overall system performance, area, power, resilience, and yield. However, as process technologies scale down to nanometer-regime geometries, the design and implementation of the embedded memory system are becoming increasingly difficult due to a number of exacerbating factors including increasing process variability, manufacturing defects, device wear out, and susceptibility to energetic particle strikes. Consequently, conventional memory resilience techniques will be unable to counter the raw bit error rate of the memory arrays in future technologies at economically feasible design points. Error correcting codes (ECC) are a widely-used and effective technique for correcting memory errors, but using conventional ECC techniques to correct more than one bit per word incurs high latency, area, and power overheads. In this work, we propose a novel ECC scheme based on erasure coding that can extend ECC to correct and detect multiple erroneous bits at low latency, area, and power overheads. Our results show that the increased memory resilience afforded by erasure-based ECC (EB-ECC) can be traded off to boost the memory performance, area, power, and yield. We show that EB-ECC, when combined with less than 5% row redundancy, can improve the cache access latency, power, and stability by over 40% on average, while maintaining near 100% yield and runtime reliability.

NAND Flash memory has been widely used for data storage due to its high density, high throughput, low cost, and low power. However, as the storage cells become smaller and with more bits programmed per cell, they are expected to suffer from reduced reliability and limited endurance. Wear-leveling and signal processing can significantly improve both reliability and endurance. However, finding optimal algorithms would require a quick and accurate characterization of Flash memory providing an insight into the error patterns. To this end, we have designed and implemented an FPGA-based framework for quick, accurate, and comprehensive characterization of Flash memories to allow efficient algorithm explorations.

SRAM design in scaled technologies requires knowledge of phenomena at the process, circuit, and architecture level. Decisions made at various levels of the design hierarchy affect the global figures of merit (FoMs) of an SRAM, such as, performance, power, area, and yield. However, the lack of a quick mechanism to understand the impact of changes at various levels of the hierarchy on global FoMs makes an accurate assessment of SRAM design innovations difficult. Thus, we introduce Virtual Prototyper (ViPro), a tool that helps SRAM designers explore the large design space by rapidly generating optimized virtual prototypes of complete SRAM macros. It does so by allowing designers to describe the SRAM components with varying levels of detail and by incorporating them into a hierarchical model that captures circuit and architectural features of the SRAM to optimize a complete prototype. It generates base-case prototypes that provide starting points for design space exploration, and assesses the impact of process, circuit, and architectural changes on the overall SRAM macro design.

A designer's intent and knowledge about the critical issues and trade-offs underlying a custom circuit design are implicit in the simulations she sets up for design creation and verification. However, this knowledge is tightly conjoined with technology-specific features and decoupled from the final schematic in traditional design flows. As a result, this knowledge is easily lost when the technology specifics change. This paper presents a technology agnostic simulation environment (TASE), which is a tool that uses simulation templates to capture the designer's knowledge and separate it from the technology-specific components of a simulation. TASE also allows the designer to form groups of related simulations and port them as a unit to a new technology. This allows an actual design schematic to remain tied to the analyses that illuminate the underlying trade-offs and design issues, unlike the case where schematics are ported alone. Giving the designer immediate access to the trade-offs, which are likely to change in new technologies, accelerates the re-design that often must accompany porting of complicated custom circuits. We demonstrate the usefulness of TASE by investigating Read and Write noise margins for a 6T SRAM in predictive technologies down to 16 nm.