Related papers: apenext: A Multi-Tflops LQCD Computing Project
Federated inference enhances LLM performance in edge computing through weighted averaging of distributed model predictions. However, autoregressive LLM inference requires frequent full-model forward passes across workers, severely limiting…
This thesis (extended abstract) presents the software development efforts toward efficient exploitation of heterogeneity through intricate mapping of computational kernels, collaborative execution of multiple processing elements and…
The deployment of the next generation computing platform at ExaFlops scale requires to solve new technological challenges mainly related to the impressive number (up to 10^6) of compute elements required. This impacts on system power…
The explosive arrival of OpenAI's ChatGPT has fueled the globalization of large language model (LLM), which consists of billions of pretrained parameters that embodies the aspects of syntax and semantics. HyperAccel introduces latency…
It is usually infeasible to fit and train an entire large deep neural network (DNN) model using a single edge device due to the limited resources. To facilitate intelligent applications across edge devices, researchers have proposed…
We consider "virtual" full-duplex relaying by means of half-duplex relays. In this configuration, each relay stage in a multi-hop relaying network is formed by at least two relays, used alternatively in transmit and receive modes, such that…
Post-training quantization (PTQ) aims to preserve model-level behavior; however, most methods focus on individual linear layers. Even recent extensions, such as QEP and LoaQ, which mitigate error propagation or target specific submodules,…
This Letter considers the design for computing facilities that are complementary to the leadership class High Performance Computing (HPC) facilities. This design envisions a future where funding agencies are allocating greater resources for…
Consistency Models (CMs) have significantly accelerated the sampling process in diffusion models, yielding impressive results in synthesizing high-resolution images. To explore and extend these advancements to point-cloud-based 3D shape…
This paper introduces QCDLAB, a design and research tool for lattice QCD algorithms. The tool, a collection of MATLAB functions, is based on a ``small-code'' and a ``minutes-run-time'' algorithmic design philosophy. The present version uses…
The rapid advancements in artificial intelligence (AI), particularly the Large Language Models (LLMs), have profoundly affected our daily work and communication forms. However, it is still a challenge to deploy LLMs on resource-constrained…
In order to follow up with mission-critical applications, new features need to be carried to satisfy a reliable communication with reduced latency. With this regard, this paper proposes a low latency cooperative transmission scheme, where…
We consider various {\em multi-vehicle versions of the minimum latency problem}. There is a fleet of $k$ vehicles located at one or more depot nodes, and we seek a collection of routes for these vehicles that visit all nodes so as to…
Convolutional neural networks (CNNs) have been widely employed in many applications such as image classification, video analysis and speech recognition. Being compute-intensive, CNN computations are mainly accelerated by GPUs with high…
The Knights Landing (KNL) is the codename for the latest generation of Intel processors based on Intel Many Integrated Core (MIC) architecture. It relies on massive thread and data parallelism, and fast on-chip memory. This processor…
FPGA accelerators for lightweight neural convolutional networks (LWCNNs) have recently attracted significant attention. Most existing LWCNN accelerators focus on single-Computing-Engine (CE) architecture with local optimization. However,…
The DEEP projects have developed a variety of hardware and software technologies aiming at improving the efficiency and usability of next generation high-performance computers. They evolve around an innovative concept for heterogeneous…
Since the development of the Beowulf project to build a parallel computer from commodity PC components, there have been many such clusters built. The MILC QCD code has been run on a variety of clusters and supercomputers. Key design…
An overview is given of the lessons learned from the introduction of multi-threading using OpenMP in tmLQCD. In particular, programming style, performance measurements, cache misses, scaling, thread distribution for hybrid codes, race…
This work is for designing one-stage lightweight detectors which perform well in terms of mAP and latency. With baseline models each of which targets on GPU and CPU respectively, various operations are applied instead of the main operations…