Related papers: Benchmarking High Bandwidth Memory on FPGAs
The growing demand for efficient long-sequence modeling on edge devices has propelled widespread adoption of State Space Models (SSMs) like Mamba, due to their superior computational efficiency and scalability. As its autoregressive…
FPGA-based heterogeneous architectures provide programmers with the ability to customize their hardware accelerators for flexible acceleration of many workloads. Nonetheless, such advantages come at the cost of sacrificing programmability.…
Current HPC systems provide memory resources that are statically configured and tightly coupled with compute nodes. However, workloads on HPC systems are evolving. Diverse workloads lead to a need for configurable memory resources to…
Modern analytics and recommendation systems are increasingly based on graph data that capture the relations between entities being analyzed. Practical graphs come in huge sizes, offer massive parallelism, and are stored in sparse-matrix…
As the size of DLRMs gets larger, the models must be partitioned across multiple GPUs or nodes of GPUs due to the size limitation of total HBM memory that can be packaged in a GPU. This partitioning adds communication and synchronization…
Many modern workloads, such as neural networks, databases, and graph processing, are fundamentally memory-bound. For such workloads, the data movement between main memory and CPU cores imposes a significant overhead in terms of both latency…
The hybrid hiding encryption algorithm, as its name implies, embraces concepts from both steganography and cryptography. In this exertion, an improved micro-architecture Field Programmable Gate Array (FPGA) implementation of this algorithm…
FPGA-based hardware accelerators are becoming increasingly popular due to their versatility, customizability, energy efficiency, constant latency, and scalability. FPGAs can be tailored to specific algorithms, enabling efficient hardware…
This article features extended summaries and retrospectives of some of the recent research done by our research group, SAFARI, on (1) various critical problems in memory systems and (2) how memory system bottlenecks affect graphics…
Ultrafast ultrasound imaging is essential for advanced ultrasound imaging techniques such as ultrasound localization microscopy (ULM) and functional ultrasound (fUS). Current ultrafast ultrasound imaging is challenged by the ultrahigh data…
When multiple processor cores (CPUs) and a GPU integrated together on the same chip share the off-chip DRAM, requests from the GPU can heavily interfere with requests from the CPUs, leading to low system performance and starvation of cores.…
Efficient on-device neural network (NN) inference offers predictable latency, improved privacy and reliability, and lower operating costs for vendors than cloud-based inference. This has sparked recent development of microcontroller-scale…
Due to the emergence of embedded applications in image and video processing, communication and cryptography, improvement of pictorial information for better human perception like deblurring, denoising in several fields such as satellite…
The emergence of high-density byte-addressable non-volatile memory (NVM) is promising to accelerate data- and compute-intensive applications. Current NVM technologies have lower performance than DRAM and, thus, are often paired with DRAM in…
This book focuses on the use of algorithmic high-level synthesis (HLS) to build application-specific FPGA systems. Our goal is to give the reader an appreciation of the process of creating an optimized hardware design using HLS. Although…
Energy consumption dictates the cost and environmental impact of deploying Large Language Models. This paper investigates the impact of on-chip SRAM size and operating frequency on the energy efficiency and performance of LLM inference,…
State-of-the-art \emph{software transactional memory (STM)} implementations achieve good performance by carefully avoiding the overhead of \emph{incremental validation} (i.e., re-reading previously read data items to avoid inconsistency)…
The rapid advancement of Large Language Models (LLMs) necessitates a deep understanding of their fundamental performance limits. This paper investigates the limits of LLM inference, focusing on hardware-imposed bottlenecks in…
Processor-in-Memory (PIM) overlays and new redesigned reconfigurable tile fabrics have been proposed to eliminate the von Neumann bottleneck and enable processing performance to scale with BRAM capacity. The performance of these FPGA-based…
Modern multicore system-on-chips (SoCs) share off-chip DRAM across cores, where bank-level interference can significantly degrade performance and threaten real-time guarantees. While prior work has focused on per-core bandwidth regulation,…