相关论文: The Ghost in the Datacenter: Link Flapping, Topolo…

This study introduces a robust solution for the detection of Distributed Denial of Service (DDoS) attacks in Internet of Things (IoT) systems, leveraging the capabilities of Graph Convolutional Networks (GCN). By conceptualizing IoT devices…

Despite the remarkable progress of diffusion models in image generation, recent studies reveal their vulnerability to backdoor attacks via covert visual or textual triggers. Although evolving defense mechanisms can detect most existing…

Large Language Models (LLMs) are increasingly deployed on converged Cloud and High-Performance Computing (HPC) infrastructure. However, as LLMs handle confidential inputs and are fine-tuned on costly, proprietary datasets, their heightened…

Tensor parallelism (TP) in large-scale LLM inference and training introduces frequent collective operations that dominate inter-GPU communication. While in-switch computing, exemplified by NVLink SHARP (NVLS), accelerates collective…

With the rapid development of large language models (LLMs), distributed training and inference frameworks like DeepSpeed have become essential for scaling model training and inference across multiple GPUs or nodes. However, the increasing…

A satisfactory understanding of information processing in spiking neural networks requires appropriate computational abstractions of neural activity. Traditionally, the neural population state vector has been the most common abstraction…

Algorithms for finding minimum or bounded vertex covers in graphs use a branch-and-reduce strategy, which involves exploring a highly imbalanced search tree. Prior GPU solutions assign different thread blocks to different sub-trees, while…

Deep Neural Networks (DNN) have found numerous applications in various domains, including fraud detection, medical diagnosis, facial recognition, and autonomous driving. However, DNN-based systems often suffer from reliability issues due to…

While GPU clusters are the de facto choice for training large deep neural network (DNN) models today, several reasons including ease of workflow, security and cost have led to efforts investigating whether CPUs may be viable for inference…

The final MLP of GPT-2 Small exhibits a fully legible routing program -- 27 named neurons organized into a three-tier exception handler -- while the knowledge it routes remains entangled across ~3,040 residual neurons. We decompose all…

Fault-tolerant routing (FTR) in Networks-on-Chip (NoCs) has become a common practice to sustain the performance of multi-core systems with an increasing number of faults on a chip. On the other hand, usage of third-party intellectual…

Generative models have achieved remarkable success across various applications, driving the demand for multi-GPU computing. Inter-GPU communication becomes a bottleneck in multi-GPU computing systems, particularly on consumer-grade GPUs. By…

Mixture-of-Experts (MoE) serving relies on wide expert parallelism (EP) to aggregate the memory capacity and bandwidth of many GPUs within one inference instance. This efficiency comes with a systems cost: every decoding step depends on…

Scaling distributed GPU training is commonly assumed to yield predictable performance gains as additional nodes are added. In practice, many large-scale deployments encounter diminishing returns and unstable behavior well before theoretical…

Access to fiber-optic connectivity in the Internet is traditionally offered either via lit circuits or dark fiber. Economic (capex vs. opex) and operational considerations (latency, capacity) dictate the choice between these two offerings,…

Distributed GPU applications increasingly rely on kernel-level, cross-node coordination to reduce launch overheads and improve compute-communication overlap, but such support is lacking. On OFI-based interconnects such as HPE Slingshot,…

Cluster identification tasks occur in a multitude of contexts in physics and engineering such as, for instance, cluster algorithms for simulating spin models, percolation simulations, segmentation problems in image processing, or network…

Deploying convolutional neural networks (CNNs) on mobile devices is difficult due to the limited memory and computation resources. We aim to design efficient neural networks for heterogeneous devices including CPU and GPU, by exploiting the…

A unique feature of neuromorphic computing is that memory is an implicit part of processing through traces of past information in the system's collective dynamics. The extent of memory about past inputs is commonly quantified by the…

Deep graph clustering, which aims to group the nodes of a graph into disjoint clusters with deep neural networks, has achieved promising progress in recent years. However, the existing methods fail to scale to the large graph with million…