Related papers: TileLang: A Composable Tiled Programming Model for…
Communication has become a first-order bottleneck in large-cale GPU workloads, and existing distributed compilers address it mainly by overlapping whole compute and communication kernels at the stream level. This coarse granularity incurs…
High-performance GPU kernel optimization remains a critical yet labor-intensive task in modern machine learning workloads. Although Triton, a domain-specific language for GPU programming, enables developers to write efficient kernels with…
Deep neural networks have recently achieved state of the art performance thanks to new training algorithms for rapid parameter estimation and new regularization methods to reduce overfitting. However, in practice the network architecture…
In-network computation represents a transformative approach to addressing the escalating demands of Artificial Intelligence (AI) workloads on network infrastructure. By leveraging the processing capabilities of network devices such as…
The accuracy and complexity of machine learning algorithms based on kernel optimization are limited by the set of kernels over which they are able to optimize. An ideal set of kernels should: admit a linear parameterization (for…
High-performance GPU kernels are critical to modern machine learning systems, yet developing efficient implementations remains a challenging, expert-driven process due to the tight coupling between algorithmic structure, memory hierarchy…
In this work, we propose KPerfIR, a novel multilevel compiler-centric infrastructure to enable the development of customizable, extendable, and portable profiling tools tailored for modern artificial intelligence (AI) workloads on modern…
In the current era of Big Data, data engineering has transformed into an essential field of study across many branches of science. Advancements in Artificial Intelligence (AI) have broadened the scope of data engineering and opened up new…
We present Nautilus, a novel tensor compiler that moves toward fully automated math-to-kernel optimization. Nautilus compiles a high-level algebraic specification of tensor operators into efficient tiled GPU kernels. Nautilus's successive…
There is a growing interest in custom spatial accelerators for machine learning applications. These accelerators employ a spatial array of processing elements (PEs) interacting via custom buffer hierarchies and networks-on-chip. The…
To overcome the well-known memory bottleneck of AI chips, 3D stacked architectures that employ advanced packaging technology with high-density through-silicon vias (TSVs) pins have proven to be a promising solution. The 3D-stacked AI chip…
The rapid growth of end-user AI applications, such as computer vision and generative AI, has led to immense data and processing demands often exceeding user devices' capabilities. Edge AI addresses this by offloading computation to the…
High-level synthesis, source-to-source compilers, and various Design Space Exploration techniques for pragma insertion have significantly improved the Quality of Results of generated designs. These tools offer benefits such as reduced…
Compound AI applications, composed from interactions between Large Language Models (LLMs), Machine Learning (ML) models, external tools and data sources are quickly becoming an integral workload in datacenters. Their diverse sub-components…
Sparse tensor algebra is a challenging class of workloads to accelerate due to low arithmetic intensity and varying sparsity patterns. Prior sparse tensor algebra accelerators have explored tiling sparse data to increase exploitable data…
Fast-evolving artificial intelligence (AI) algorithms such as large language models have been driving the ever-increasing computing demands in today's data centers. Heterogeneous computing with domain-specific architectures (DSAs) brings…
Interest in applying Artificial Intelligence (AI) techniques to compiler optimizations is increasing rapidly, but compiler research has a high entry barrier. Unlike in other domains, compiler and AI researchers do not have access to the…
Writing high-performance code requires significant expertise in the programming language, compiler optimizations, and hardware knowledge. This often leads to poor productivity and portability and is inconvenient for a non-programmer…
Improving GPU kernel efficiency is crucial for advancing AI systems. Recent work has explored leveraging large language models (LLMs) for GPU kernel generation and optimization. However, existing LLM-based kernel optimization pipelines…
Neural networks can be regarded as a new programming paradigm, i.e., instead of building ever-more complex programs through (often informal) logical reasoning in the programmers' mind, complex 'AI' systems are built by optimising generic…