Related papers: LEO: Tracing GPU Stall Root Causes via Cross-Vendo…
Achieving high performance for GPU codes requires developers to have significant knowledge in parallel programming and GPU architectures, and in-depth understanding of the application. This combination makes it challenging to find…
While Model Predictive Control (MPC) delivers strong performance across robotics applications, solving the underlying (batches of) nonlinear trajectory optimization (TO) problems online remains computationally demanding. Existing…
Analyzing IO performance anomalies is a crucial task in various computing environments, ranging from large-scale cloud applications to desktop applications. However, the IO stack of modern operating systems is complicated, making it hard to…
Detecting performance issues and identifying their root causes in the runtime is a challenging task. Typically, developers use methods such as logging and tracing to identify bottlenecks. These solutions are, however, not ideal as they are…
GPUs are the most popular platform for accelerating HPC workloads, such as artificial intelligence and science simulations. However, most microarchitectural research in academia relies on GPU core pipeline designs based on architectures…
Parallel accelerators, such as GPUs, are key enablers for large-scale Machine Learning (ML) applications. However, ML model developers often lack detailed knowledge of the underlying system architectures, while system programmers usually do…
Fail-slows, or stragglers, are common but largely unheeded problems in large-scale hybrid-parallel training that spans thousands of GPU servers and runs for weeks to months. Yet, these problems are not well studied, nor can they be quickly…
Large-scale GPU traces play a critical role in identifying performance bottlenecks within heterogeneous High-Performance Computing (HPC) architectures. However, the sheer volume and complexity of a single trace of data make performance…
We present DIO, a generic tool for observing inefficient and erroneous I/O interactions between applications and in-kernel storage systems that lead to performance, dependability, and correctness issues. DIO facilitates the analysis and…
Disk-based graph indexes for approximate nearest neighbor search (ANNS) must serve latency-sensitive queries and throughput-demanding updates concurrently. We observe that over 40% of search-thread CPU time is spent stalling on disk I/O;…
Searching for sources of electromagnetic emission in spectral-line radio astronomy interferometric data is a computationally intensive process. Parallel programming techniques and High Performance Computing hardware may be used to improve…
GPUs are a key enabler of the revolution in machine learning and high performance computing, functioning as de facto co-processors to accelerate large-scale computation. As the programming stack and tool support have matured, GPUs have also…
We address the problem of learning observation models end-to-end for estimation. Robots operating in partially observable environments must infer latent states from multiple sensory inputs using observation models that capture the joint…
Developing efficient GPU kernels can be difficult because of the complexity of GPU architectures and programming models. Existing performance tools only provide coarse-grained suggestions at the kernel level, if any. In this paper, we…
Core decomposition is a well-established graph mining problem with various applications that involves partitioning the graph into hierarchical subgraphs. Solutions to this problem have been developed using both bottom-up and top-down…
Optimizing GPU kernels with LLM agents is an iterative process over a large design space. Every candidate must be generated, compiled, validated, and profiled, so fewer trials will save both runtime and cost. We make two key observations.…
Graphics processors, or GPUs, have recently been widely used as accelerators in the shared environments such as clusters and clouds. In such shared environments, many kernels are submitted to GPUs from different users, and throughput is an…
Large-scale AI training is now fundamentally a distributed systems problem, and hardware failures have become routine operating conditions rather than rare exceptions. Public operational evidence from production training clusters, however,…
Large industrial systems that combine services and applications, have become targets for cyber criminals and are challenging from the security, monitoring and auditing perspectives. Security log analysis is a key step for uncovering…
We propose an optimization approach for determining both hardware and software parameters for the efficient implementation of a (family of) applications called dense stencil computations on programmable GPGPUs. We first introduce a simple,…