Related papers: UB-Mesh: a Hierarchically Localized nD-FullMesh Da…
Mobile Edge Computing (MEC) reduces the computational burden on terminal devices by shortening the distance between these devices and computing nodes. Integrating Unmanned Aerial Vehicles (UAVs) with enhanced MEC networks can leverage the…
We propose DFModel, a modeling framework for mapping dataflow computation graphs onto large-scale systems. Mapping a workload to a system requires optimizing dataflow mappings at various levels, including the inter-chip (between chips)…
Unmanned aerial vehicles (UAVs) play significant roles in multiple fields, which brings great challenges for the airspace safety. In order to achieve efficient surveillance and break the limitation of application scenarios caused by single…
This paper presents a predictive deep learning framework for dynamic sub-band allocation in Sub-Band Full Duplex (SBFD) systems, addressing the challenge of balancing uplink (UL) and downlink (DL) performance under highly dynamic traffic…
Distributed training is a solution to reduce DNN training time by splitting the task across multiple NPUs (e.g., GPU/TPU). However, distributed training adds communication overhead between the NPUs in order to synchronize the gradients…
The rise of delay-sensitive yet computing-intensive Internet of Things (IoT) applications poses challenges due to the limited processing power of IoT devices. Mobile Edge Computing (MEC) offers a promising solution to address these…
The scaling law for large language models (LLMs) depicts that the path towards machine intelligence necessitates training at large scale. Thus, companies continuously build large-scale GPU clusters, and launch training jobs that span over…
The growing demand for multi-DNN workloads with unpredictable task arrival times has highlighted the need for interruptible scheduling on edge accelerators. However, existing preemptive frameworks typically assume known task arrival times…
Unified Multimodal Models (UMMs) built on shared autoregressive (AR) transformers are attractive for their architectural simplicity. However, we identify a critical limitation: when trained on multimodal inputs, modality-shared transformers…
Efficient relocalization is essential for intelligent vehicles when GPS reception is insufficient or sensor-based localization fails. Recent advances in Bird's-Eye-View (BEV) segmentation allow for accurate estimation of local scene…
Large Language Models (LLMs) have revolutionized Artificial Intelligence (AI) with significant advancements such as OpenAI's ChatGPT, Meta's Llama, and Databricks' DBRX. This paper addresses the cost and scalability challenges encountered…
We show how to quantify scalability with the Universal Scalability Law (USL) by applying it to performance measurements of memcached, J2EE, and Weblogic on multi-core platforms. Since commercial multicores are essentially black-boxes, the…
Modern Machine Learning (ML) training on large-scale datasets is a very time-consuming workload. It relies on the optimization algorithm Stochastic Gradient Descent (SGD) due to its effectiveness, simplicity, and generalization performance.…
With the growing demand for latency-critical and computation-intensive Internet of Things (IoT) services, the IoT-oriented network architecture, mobile edge computing (MEC), has emerged as a promising technique to reinforce the computation…
Full-duplex (FD) wireless is an attractive communication paradigm with high potential for improving network capacity and reducing delay in wireless networks. Despite significant progress on the physical layer development, the challenges…
Training large language models (LLMs) in the cloud faces growing memory bottlenecks due to the limited capacity and high cost of GPUs. While GPU memory offloading to CPU and NVMe has made large-scale training more feasible, existing…
Large language models (LLMs) such as OpenAI's ChatGPT and Google's Gemini have demonstrated unprecedented capabilities of autoregressive AI models across multiple tasks triggering disruptive technology innovations around the world. However,…
Data and pipeline parallelism are key strategies for scaling neural network training across distributed devices, but their high communication cost necessitates co-located computing clusters with fast interconnects, limiting their…
Ultra-dense networks (UDNs) represent a transformative access architecture for upcoming sixth generation (6G) systems, poised to meet the surging demand for high data rates. Achieving precise synchronization across diverse base stations…
Edge deployment of low-batch large language models (LLMs) faces critical memory bandwidth bottlenecks when executing memory-intensive general matrix-vector multiplications (GEMV) operations. While digital processing-in-memory (PIM)…