Related papers: RNTuple: Towards First-Class Support for HPC data …
The Chain-of-Thought (CoT) paradigm, while enhancing the interpretability of Large Language Models (LLMs), is constrained by the inefficiencies and expressive limits of natural language. Latent Chain-of-Thought (latent CoT) reasoning, which…
Datacenters are the backbone of our digital society, but raise numerous operational challenges. We envision digital twins becoming primary instruments in datacenter operations, continuously and autonomously helping with major operational…
Lossy compression is one of the most efficient solutions to reduce storage overhead and improve I/O performance for HPC applications. However, existing parallel I/O libraries cannot fully utilize lossy compression to accelerate parallel…
The scaling of large language models (LLMs) is currently bottlenecked by the rigidity of distributed programming. While high-performance libraries like CuBLAS and NCCL provide optimized primitives, they lack the flexibility required for…
This paper introduces DeepCircuitX, a comprehensive repository-level dataset designed to advance RTL (Register Transfer Level) code understanding, generation, and power-performance-area (PPA) analysis. Unlike existing datasets that are…
We propose the Ratio1 AI meta-operating system (meta-OS), a decentralized MLOps protocol that unifies AI model development, deployment, and inference across heterogeneous edge devices. Its key innovation is an integrated blockchain-based…
Heterogeneous systems have become one of the most common architectures today, thanks to their excellent performance and energy consumption. However, due to their heterogeneity they are very complex to program and even more to achieve…
Managing and preparing complex data for deep learning, a prevalent approach in large-scale data science can be challenging. Data transfer for model training also presents difficulties, impacting scientific fields like genomics, climate…
During the years 2000 and 2001 the HERA machine and the H1 experiment performed substantial luminosity upgrades. To cope with the increased demands on data handling an effort was made to redesign and modernize the analysis software. Main…
The pervasive adoption of Deep Learning (DL) and Graph Processing (GP) makes it a de facto requirement to build large-scale clusters of heterogeneous accelerators including GPUs and FPGAs. The OpenCL programming framework can be used on the…
The performance of large-scale computing systems often critically depends on high-performance communication networks. Dynamically reconfigurable topologies, e.g., based on optical circuit switches, are emerging as an innovative new…
The significant resource demands in LLM serving prompts production clusters to fully utilize heterogeneous hardware by partitioning LLM models across a mix of high-end and low-end GPUs. However, existing parallelization approaches often…
The growing scale of deep learning demands distributed training frameworks that jointly reason about parallelism, memory, and network topology. Prior works often rely on heuristic or topology-agnostic search, handling communication and…
The explosive growth of Large Language Models (LLMs), such as GPT-4 with 1.8 trillion parameters, demands a fundamental rethinking of data center architecture to ensure scalability, efficiency, and cost-effectiveness. Our work provides a…
Inspired by advances in LLMs, reasoning-enhanced sequential recommendation performs multi-step deliberation before making final predictions, unlocking greater potential for capturing user preferences. However, current methods are…
The rapid advancement of embedded multicore and many-core systems has revolutionized computing, enabling the development of high-performance, energy-efficient solutions for a wide range of applications. As models scale up in size, data…
Many edge devices employ Recurrent Neural Networks (RNN) to enhance their product intelligence. However, the increasing computation complexity poses challenges for performance, energy efficiency and product development time. In this paper,…
The High-Throughput Experimental Materials Database (HTEM-DB) is the endpoint repository for inorganic thin-film materials data collected during combinatorial experiments at the National Renewable Energy Laboratory (NREL). This unique data…
Parameter-efficient transfer learning (PETL), i.e., fine-tuning a small portion of parameters, is an effective strategy for adapting pre-trained models to downstream domains. To further reduce the memory demand, recent PETL works focus on…
I/O efficiency is crucial to productivity in scientific computing, but the increasing complexity of the system and the applications makes it difficult for practitioners to understand and optimize I/O behavior at scale. Data-driven machine…