Related papers: Fake Runs, Real Fixes -- Analyzing xPU Performance…

Benchmarking and co-design are essential for driving optimizations and innovation around ML models, ML software, and next-generation hardware. Full workload benchmarks, e.g. MLPerf, play an essential role in enabling fair comparison across…

Growing deployment of power and energy efficient throughput accelerators (GPU) in data centers demands enhancement of power-performance co-optimization capabilities of GPUs. Realization of exascale computing using accelerators requires…

Stochastic simulation techniques employed for the analysis of portfolios of insurance/reinsurance risk, often referred to as `Aggregate Risk Analysis', can benefit from exploiting state-of-the-art high-performance computing platforms. In…

Evaluative claims about LLM infrastructure -- ``workload X is fastest on hardware Y with software Z'' -- depend on a complex configuration space spanning hardware accelerators, interconnect bandwidth, software frameworks, parallelism plans,…

We explore the idea of integrating machine learning (ML) with high performance computing (HPC)-driven simulations to address challenges in using simulations to teach computational science and engineering courses. We demonstrate that a ML…

Large language models~(LLMs) are known for their high demand on computing resources and memory due to their substantial model size, which leads to inefficient inference on moderate GPU systems. Techniques like quantization or pruning can…

In recent decades, High Performance Computing (HPC) has undergone significant enhancements, particularly in the realm of hardware platforms, aimed at delivering increased processing power while keeping power consumption within reasonable…

Accurate simulation techniques are indispensable to efficiently propose new memory or architectural organizations. As implementing new hardware concepts in real systems is often not feasible, cycle-accurate simulators employed together with…

This paper presents a scheme for efficient channel usage between simulator and accelerator where the accelerator models some RTL sub-blocks in the accelerator-based hardware/software co-simulation while the simulator runs transaction-level…

Machine learning (ML) involves private data and proprietary model parameters. MPC-based ML allows multiple parties to collaboratively run an ML workload without sharing their private data or model parameters using multi-party computing…

Due to the cost-prohibitive nature of training Large Language Models (LLMs), fine-tuning has emerged as an attractive alternative for specializing LLMs for specific tasks using limited compute resources in a cost-effective manner. In this…

Modern AI accelerators rely on matrix multiply-accumulate units (MMAUs), such as NVIDIA Tensor Cores and AMD Matrix Cores, to accelerate deep neural network workloads. MMAUs expose only instruction-level or API-level interfaces of matrix…

Split-execution computing leverages the capabilities of multiple computational models to solve problems, but splitting program execution across different computational models incurs costs associated with the translation between domains. We…

Nowadays, GPU accelerators are commonly used to speed up general-purpose computing tasks on a variety of hardware. However, due to the diversity of GPU architectures and processed data, optimization of codes for a particular type of…

Many modern parallel computing systems are heterogeneous at their node level. Such nodes may comprise general purpose CPUs and accelerators (such as, GPU, or Intel Xeon Phi) that provide high performance with suitable energy-consumption…

Sustaining exascale performance in production requires engineering choices and operational practices that emerge only under real deployment constraints and demand coordination across system layers. This paper reports experience from three…

FPGAs have found increasing adoption in data center applications since a new generation of high-level tools have become available which noticeably reduce development time for FPGA accelerators and still provide high quality of results.…

The evolution of distributed architectures and programming paradigms for performance-oriented program development, challenge the state-of-the-art technology for performance tools. The area of high performance computing is rapidly expanding…

Training and deploying large-scale machine learning models is time-consuming, requires significant distributed computing infrastructures, and incurs high operational costs. Our analysis, grounded in real-world large model training on…

The rapid growth of large-language models (LLMs) is driving a new wave of specialized hardware for inference. This paper presents the first workload-centric, cross-architectural performance study of commercial AI accelerators, spanning…