Related papers: Morphlux: Transforming Torus Fabrics for Efficient…

Recently, transformer-based approaches have shown promising results for semi-supervised video object segmentation. However, these approaches typically struggle on long videos due to increased GPU memory demands, as they frequently expand…

Deploying a large language model (LLM) inference service remains costly because centralized serving depends on specialized GPU clusters and high-bandwidth interconnects in datacenters. An appealing alternative is to leverage collaborative…

The resurgence of machine learning has increased the demand for high-performance basic linear algebra subroutines (BLAS), which have long depended on libraries to achieve peak performance on commodity hardware. High-performance BLAS…

Data center hardware refresh cycles are lengthening. However, increasing processor complexity is raising the potential for faults. To achieve longevity in the face of increasingly fault-prone datapaths, fault tolerance is needed, especially…

High-multiplicity events remain a bottleneck for LHC simulations due to their computational cost. We present a ML-surrogate approach to accelerate matrix element reweighting from leading-color (LC) to full-color (FC) accuracy, building on…

Recent breakthroughs in Large-scale language models (LLMs) have demonstrated impressive performance on various tasks. The immense sizes of LLMs have led to very high resource demand and cost for running the models. Though the models are…

Performance and scalability are major concerns for blockchains: permissionless systems are typically limited by slow proof of X consensus algorithms and sequential post-order transaction execution on every node of the network. By…

While Transformers have achieved remarkable success in LLMs through superior scalability, their application in industrial-scale ranking models remains nascent, hindered by the challenges of high feature sparsity and low label density. In…

In Polaris, we introduced a cloud-native distributed query processor to perform analytics at scale. In this paper, we extend the underlying Polaris distributed computation framework, which can be thought of as a read-only transaction…

Collaborative Machine Learning is a paradigm in the field of distributed machine learning, designed to address the challenges of data privacy, communication overhead, and model heterogeneity. There have been significant advancements in…

The Transformer Machine Learning (ML) architecture has been gaining considerable momentum in recent years. In particular, computational High-Energy Physics tasks such as jet tagging and particle track reconstruction (tracking), have either…

The integration of dynamic, sparse structures like Mixture-of-Experts (MoE) with parameter-efficient adapters (e.g., LoRA) is a powerful technique for enhancing Large Language Models (LLMs). However, this architectural enhancement comes at…

In this work, we propose a configurable many-core overlay for high-performance embedded computing. The size of internal memory, supported operations and number of ports can be configured independently for each core of the overlay. The…

Distributed training frameworks, like TensorFlow, have been proposed as a means to reduce the training time of deep learning models by using a cluster of GPU servers. While such speedups are often desirable---e.g., for rapidly evaluating…

Hyperledger Fabric (HLF) is a secure and robust blockchain (BC) platform that supports high-throughput and low-latency transactions. However, it encounters challenges in managing conflicting transactions that negatively affect throughput…

Large language models (LLMs) often face a bottleneck in inference speed due to their reliance on auto-regressive decoding. Recently, parallel decoding has shown significant promise in enhancing inference efficiency. However, we have…

We propose the CLEX supercomputer topology and routing scheme. We prove that CLEX can utilize a constant fraction of the total bandwidth for point-to-point communication, at delays proportional to the sum of the number of intermediate hops…

Serverless computing has grown rapidly for serving Large Language Model (LLM) inference due to its pay-as-you-go pricing, fine-grained GPU usage, and rapid scaling. However, our analysis reveals that current serverless can effectively serve…

This paper presents a broad, pathfinding design space exploration of memory management units (MMUs) for heterogeneous systems. We consider a variety of designs, ranging from accelerators tightly coupled with CPUs (and using their MMUs) to…

Finding the convex hull is a fundamental problem in computational geometry. Quickhull is a fast algorithm for finding convex hulls. In this paper, we present VQhull, a fast parallel implementation of Quickhull that exploits vector…