Related papers: A Context-Switching/Dual-Context ROM Augmented RAM…
There is growing demand for performing inference with hundreds of thousands of input tokens on trained transformer models. Inference at this extreme scale demands significant computational resources, hindering the application of…
Performing data-intensive tasks in the von Neumann architecture is challenging to achieve both high performance and power efficiency due to the memory wall bottleneck. Computing-in-memory (CiM) is a promising mitigation approach by enabling…
As conventional memory technologies are challenged by their technological physical limits, emerging technologies driven by novel materials are becoming an attractive option for future memory architectures. Among these technologies,…
In this work, we present a novel non-volatile spin transfer torque (STT) assisted spin-orbit torque (SOT) based ternary content addressable memory (TCAM) with 5 transistors and 2 magnetic tunnel junctions (MTJs). We perform a comprehensive…
Resistance switching devices are of special importance because of their application in resistive memories (RRAM) which are promising candidates for replacing current nonvolatile memories and realize storage class memories. These devices…
Transformer-based models have become ubiquitous in natural language processing thanks to their large capacity, innate parallelism and high performance. The contextualizing component of a Transformer block is the $\textit{pairwise…
Retrieval-augmented generation (RAG) enhances large language models (LLMs) with external knowledge but incurs significant inference costs due to lengthy retrieved contexts. While context compression mitigates this issue, existing methods…
Byte-addressable persistent memory (B-APM) presents a new opportunity to bridge the performance gap between main memory and storage. In this paper, we present the usage scenarios for this new technology, based on the capabilities of Intel's…
Data copy is a widely-used memory operation in many programs and operating system services. In conventional computers, data copy is often carried out by two separate read and write transactions that pass data back and forth between the DRAM…
The latest trends in high-performance computing systems show an increasing demand on the use of a large scale multicore systems in a efficient way, so that high compute-intensive applications can be executed reasonably well. However, the…
Dynamic Random Access Memory (DRAM) is the de-facto choice for main memory devices due to its cost-effectiveness. It offers a larger capacity and higher bandwidth compared to SRAM but is slower than the latter. With each passing generation,…
Phase-change memory (PCM) devices have multiple banks to serve memory requests in parallel. Unfortunately, if two requests go to the same bank, they have to be served one after another, leading to lower system performance. We observe that a…
LLM agents with persistent memory store information as flat factual records, providing little context for temporal reasoning, change tracking, or cross-session aggregation. Inspired by the drawing effect [3], we introduce dual-trace memory…
A content-addressable-memory compares an input search word against all rows of stored words in an array in a highly parallel manner. While supplying a very powerful functionality for many applications in pattern matching and search, it…
DRAM-based main memory and its associated components increasingly account for a significant portion of application performance bottlenecks and power budget demands inside the computing ecosystem. To alleviate the problems of storage density…
The continuing advancement of memory technology has not only fueled a surge in performance, but also substantially exacerbate reliability challenges. Traditional solutions have primarily focused on improving the efficiency of protection…
CMOS technology and its continuous scaling have made electronics and computers accessible and affordable for almost everyone on the globe; in addition, they have enabled the solutions of a wide range of societal problems and applications.…
The substantial memory requirements of Large Language Models (LLMs), particularly for long-context fine-tuning, have renewed interest in CPU offloading to augment limited GPU memory. However, as context lengths grow, relying on CPU memory…
This paper addresses the challenge of creating a neural architecture for very long sequences that requires constant time for processing new information at each time step. Our approach, Associative Recurrent Memory Transformer (ARMT), is…
This paper presents a novel architecture utilizing a 10T SRAM cell for XNOR-based in-memory computing, aimed at mitigating the extensive routing challenges typically encountered in conventional in-memory computing systems. By integrating a…