Related papers: Leveraging Architectural Support of Three Page Siz…

Large scale-free graphs are famously difficult to process efficiently: the skewed vertex degree distribution makes it difficult to obtain balanced partitioning. Our research instead aims to turn this into an advantage by partitioning the…

The RISC-V SVNAPOT Extension aims to remedy the performance overhead of the Memory Management Unit (MMU), under heavy memory loads. The Privileged Specification defines additional Natural-Power-of-Two (NAPOT) multiples of the 4KB base page…

Multi-tiered large memory systems call for rethinking of memory profiling and migration because of the unique problems unseen in the traditional memory systems with smaller capacity and fewer tiers. We develop MTM, an…

Memory latency, bandwidth, capacity, and energy increasingly limit performance. In this paper, we reconsider proposed system architectures that consist of huge (many-terabyte to petabyte scale) memories shared among large numbers of CPUs.…

Recent advances in 1-bit Large Language Models (LLMs), such as BitNet and BitNet b1.58, present a promising approach to enhancing the efficiency of LLMs in terms of speed and energy consumption. These developments also enable local LLM…

Large language models (LLMs) are increasingly used across research and industry applications, yet their inference efficiency remains a significant challenge. As the computational power of modern GPU architectures continuously improves,…

Large Language Models (LLMs) are typically trained in two phases: pre-training on large internet-scale datasets, and fine-tuning for downstream tasks. Given the higher computational demand of pre-training, it's intuitive to assume that…

We revisit large kernel design in modern convolutional neural networks (CNNs). Inspired by recent advances in vision transformers (ViTs), in this paper, we demonstrate that using a few large convolutional kernels instead of a stack of small…

Memory management operations that modify page-tables, typically performed during memory allocation/deallocation, are infamous for their poor performance in highly threaded applications, largely due to process-wide TLB shootdowns that the OS…

Binary Neural Networks (BNNs) are showing tremendous success on realistic image classification tasks. Notably, their accuracy is similar to the state-of-the-art accuracy obtained by full-precision models tailored to edge devices. In this…

Various state-of-the-art automated reasoning (AR) tools are widely used as backend tools in research of knowledge representation and reasoning as well as in industrial applications. In testing and verification, those tools often run…

Software managed byte-addressable hybrid memory systems consisting of DRAMs and NVMMs offer a lot of flexibility to design efficient large scale data processing applications. Operating systems (OS) play an important role in enabling the…

Large language models (LLMs) have demonstrated exceptional performance across a variety of tasks. However, their substantial scale leads to significant computational resource consumption during inference, resulting in high costs.…

This paper explores practical aspects of using a high-level functional language for GPU-based arithmetic on ``midsize'' integers. By this we mean integers of up to about a quarter million bits, which is sufficient for most practical…

Caching (also known as paging) is a classical problem concerning page replacement policies in two-level memory systems. General caching is the variant with pages of different sizes and fault costs. We give the first NP-hardness result for…

There has been significant recent interest in parallel graph processing due to the need to quickly analyze the large graphs available today. Many graph codes have been designed for distributed memory or external memory. However, today even…

Stencil computations are a key class of applications, widely used in the scientific computing community, and a class that has particularly benefited from performance improvements on architectures with high memory bandwidth. Unfortunately,…

Moving data from CERN to Pasadena at a gigabyte per second using the next generation Internet requires good networking and good disk IO. Ten Gbps Ethernet and OC192 links are in place, so now it is simply a matter of programming. This…

It is often said that one of the biggest limitations on computer performance is memory bandwidth (i.e."the memory wall problem"). In this position paper, I argue that if historical trends in computing evolution (where growth in available…

Simple economic and performance arguments suggest appropriate lifetimes for main memory pages and suggest optimal page sizes. The fundamental tradeoffs are the prices and bandwidths of RAMs and disks. The analysis indicates that with…