Related papers: FlexLink: Boosting your NVLink Bandwidth by 27% wi…

Transformers, while revolutionary, face challenges due to their demanding computational cost and large data movement. To address this, we propose HyFlexPIM, a novel mixed-signal processing-in-memory (PIM) accelerator for inference that…

Large language models (LLMs) have emerged as a powerful foundation for intelligent reasoning and decision-making, demonstrating substantial impact across a wide range of domains and applications. However, their massive parameter scales and…

We describe KVLink, an approach for efficient key-value (KV) cache reuse in large language models (LLMs). In many LLM applications, different inputs can share overlapping context, such as the same retrieved document appearing in multiple…

Dense Multi-GPU systems have recently gained a lot of attention in the HPC arena. Traditionally, MPI runtimes have been primarily designed for clusters with a large number of nodes. However, with the advent of MPI+CUDA applications and…

Recent advances in mixture-of-experts architectures have shown that individual experts models can be trained federatedly, i.e., in isolation from other experts by using a common base model to facilitate coordination. However, we hypothesize…

Transformer-based Large Language Models (LLMs) have made a significant impact on various domains. However, LLMs' efficiency suffers from both heavy computation and memory overheads. Compression techniques like sparsification and…

The evolving explosion in high data rate services and applications will soon require the use of untapped, abundant unregulated spectrum of the visible light for communications to adequately meet the demands of the fifth-generation (5G)…

Emerging Large Language Model (LLM) system patterns, such as disaggregated inference, Mixture-of-Experts (MoE) routing, and asynchronous reinforcement fine-tuning, require flexible point-to-point communication beyond simple collectives.…

Large inter-GPU all-reduce operations, prevalent throughout deep learning, are bottlenecked by communication costs. Emerging heterogeneous architectures are comprised of complex nodes, often containing $4$ GPUs and dozens to hundreds of CPU…

Neural networks have become dominant computational workloads across cloud and edge platforms, but their rapid growth in model size and deployment diversity has exposed hardware bottlenecks increasingly dominated by memory movement,…

FPGAs are increasingly prevalent in cloud deployments, serving as Smart NICs or network-attached accelerators. Despite their potential, developing distributed FPGA-accelerated applications remains cumbersome due to the lack of appropriate…

Device-side Large Language Models (LLMs) have witnessed explosive growth, offering higher privacy and availability compared to cloud-side LLMs. During LLM inference, both model weights and user data are valuable, and attackers may even…

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…

The Compute Express Link (CXL) is an open industry-standard interconnect between processors and devices such as accelerators, memory buffers, smart network interfaces, persistent memory, and solid-state drives. CXL offers coherency and…

Current multimodal large lanauge models possess strong perceptual and reasoning capabilities, however high computational and memory requirements make them difficult to deploy directly on on-device environments. While small-parameter models…

Aligning future system design with the ever-increasing compute needs of large language models (LLMs) is undoubtedly an important problem in today's world. Here, we propose a general performance modeling methodology and workload analysis of…

Orthogonal Frequency Division Multiple Access (OFDMA) as well as other orthogonal multiple access techniques fail to achieve the system capacity limit in the uplink due to the exclusivity in resource allocation. This issue is more prominent…

The aspiration of the Vision-and-Language Navigation (VLN) task has long been to develop an embodied agent with robust adaptability, capable of seamlessly transferring its navigation capabilities across various tasks. Despite remarkable…

An important ingredient of the future 5G systems will be Ultra-Reliable Low-Latency Communication (URLLC). A way to offer URLLC without intervention in the baseband/PHY layer design is to use interface diversity and integrate multiple…

In frequency division duplexing (FDD) massive multiple-input multiple-output (MIMO) systems, downlink channel state information (CSI) plays a crucial role in achieving high spectrum and energy efficiency. However, the CSI feedback overhead…