Related papers: Specifying and Testing GPU Workgroup Progress Mode…
Modern machine learning workloads such as large language model training, fine-tuning jobs are highly distributed and span across hundreds of systems with multiple GPUs. Job completion time for these workloads is the artifact of the…
In recent years, high performance scientific computing on graphics processing units (GPUs) have gained widespread acceptance. These devices are designed to offer massively parallel threads for running code with general purpose. There are…
We describe a universal modeling approach for predicting single- and multicore runtime of steady-state loops on server processors. To this end we strictly differentiate between application and machine models: An application model comprises…
GPUs are widely used to accelerate many important classes of workloads today. However, we observe that several important emerging classes of workloads, including simulation engines for deep reinforcement learning and dynamic neural…
GPU architectures have become popular for executing general-purpose programs. Their many-core architecture supports a large number of threads that run concurrently to hide the latency among dependent instructions. In modern GPU…
Recent advancements in speculative decoding have demonstrated considerable speedup across a wide array of large language model (LLM) tasks. Speculative decoding inherently relies on sacrificing extra memory allocations to generate several…
We investigate the problem of certifying optimality for sparse generalized linear models (GLMs), where sparsity is enforced through a cardinality constraint. While Branch-and-Bound (BnB) frameworks can certify optimality using perspective…
Determining whether a program terminates is a central problem in computer science. Turing's Halting Problem established termination as undecidable, showing that no algorithm can universally determine termination for all programs and inputs.…
GPU activity prediction is an important and complex problem. This is due to the high level of contention among thousands of parallel threads. This problem was mostly addressed using heuristics. We propose a representation learning approach…
Parallel algorithms on CPU and GPU are implemented for the Unified Gas-Kinetic Scheme and their performances are investigated and compared by a two dimensional channel flow case. The parallel CPU algorithm has a one dimensional block…
Global protocol specifications are the starting point of top-down verification methodologies, and serve as a blueprint for synthesizing local specifications that guarantee the correctness of distributed implementations. In this work, we…
Matlab is very widely used in scientific computing, but Matlab computational efficiency is lower than C language program. In order to improve the computing speed, some toolbox can use GPU to accelerate the computation. This paper describes…
We propose a language and compiler to productively build high-performance {\it software systolic arrays} that run on GPUs. Based on a rigorous mathematical foundation (uniform recurrence equations and space-time transform), our language has…
GPUs are readily available in cloud computing and personal devices, but their use for data processing acceleration has been slowed down by their limited integration with common programming languages such as Python or Java. Moreover, using…
On the way to Exascale, programmers face the increasing challenge of having to support multiple hardware architectures from the same code base. At the same time, portability of code and performance are increasingly difficult to achieve as…
Leveraging Graphics Processing Units (GPUs) to accelerate scientific software has proven to be highly successful, but in order to extract more performance, GPU programmers must overcome the high latency costs associated with their use. One…
The common assumption in on-device AI is that GPUs, with their superior parallel processing, always provide the best performance for large language model (LLM) inference. In this work, we challenge this notion by empirically demonstrating…
To address the challenge of performance analysis on the US DOE's forthcoming exascale supercomputers, Rice University has been extending its HPCToolkit performance tools to support measurement and analysis of GPU-accelerated applications.…
Prior work on Automatically Scalable Computation (ASC) suggests that it is possible to parallelize sequential computation by building a model of whole-program execution, using that model to predict future computations, and then…
The continued growth of the computational capability of throughput processors has made throughput processors the platform of choice for a wide variety of high performance computing applications. Graphics Processing Units (GPUs) are a prime…