On the Thermal Vulnerability of 3D-Stacked High-Bandwidth Memory Architectures
By Mehdi Elahi 1, Mohamed R. Elshamy 2, Abdel-Hameed A. Badawy 2 and Ahmad Patooghy 1
1 North Carolina A&T State University
2 New Mexico State University

Abstract
3D-stacked High Bandwidth Memory (HBM) architectures provide high-performance memory interactions to address the well-known performance challenge, namely the memory wall. However, these architectures are susceptible to thermal vulnerabilities due to the inherent vertical adjacency that occurs during the manufacturing process of HBM architectures. We anticipate that adversaries may exploit the intense vertical and lateral adjacency to design and develop thermal performance degradation attacks on the memory banks that host data/instructions from victim applications. In such attacks, the adversary manages to inject short and intense heat pulses from vertically and/or laterally adjacent memory banks, creating a convergent thermal wave that maximizes impact and delays the victim application from accessing its data/instructions. As the attacking application does not access any out-of-range memory locations, it can bypass both design-time security tests and the operating system's memory management policies. In other words, since the attack mimics legitimate workloads, it will be challenging to detect.
Keywords: 3D-Stacked Memory, HBM, Thermal Attack, Thermal Coupling, Security
To read the full article, click here
Related Semiconductor IP
- HBM 4 Verification IP
- Verification IP for HBM
- Simulation VIP for HBM
- HBM Synthesizable Transactor
- HBM DFI Synthesizable Transactor
Related Articles
- Automotive Architectures: Domain, Zonal and the Rise of Central
- SoCs: DSP World, Cores -> New DSP architectures work harder
- Processor forum examines embedded cache, architectures
- What is the impact of streaming data on SoC architectures?
Latest Articles
- RISC-V Functional Safety for Autonomous Automotive Systems: An Analytical Framework and Research Roadmap for ML-Assisted Certification
- Emulation-based System-on-Chip Security Verification: Challenges and Opportunities
- A 129FPS Full HD Real-Time Accelerator for 3D Gaussian Splatting
- SkipOPU: An FPGA-based Overlay Processor for Large Language Models with Dynamically Allocated Computation
- TensorPool: A 3D-Stacked 8.4TFLOPS/4.3W Many-Core Domain-Specific Processor for AI-Native Radio Access Networks