Mat2Boundary treats boundary conditions as sparse matrix-vector products and uses multi-stage compilation with polyhedral analysis to generate efficient matrix-free kernels and communication schedules for distributed block-structured PDE solvers.
Exocompilation for productive programming of hardware accelerators
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
roles
background 1polarities
background 1representative citing papers
A new partitioning algorithm that provably load-balances arbitrary sparse tensor algebra expressions by generalizing parallel merging to multi-operand, multi-dimensional hierarchical structures, implemented in a compiler framework.
ATLAAS automatically converts RTL-extracted bit-level accelerator semantics into tensor-level ISA specs via an 8-pass MLIR pipeline, enabling automated compiler backend generation for designs like Gemmini and VTA.
Scion is a new DSL and compiler that decouples BVH data layouts from traversal algorithms, enabling architecture-agnostic layout optimizations and a novel Pareto-optimal ray tracing layout.
NESA presents a neuro-symbolic framework that decomposes static analyses into policy-defined sub-problems solved by parsers and LLMs to enable compilation-free customizable analysis with reduced hallucinations.
EDGE extends Einsum notation with graph-specific operations to create a unified tensor-algebra framework for expressing and manipulating graph algorithms.
citing papers explorer
-
Mat2Boundary: Treating User-Defined Boundary Condition as SpMV for Distributed PDE Solvers on Block-Structured Grids
Mat2Boundary treats boundary conditions as sparse matrix-vector products and uses multi-stage compilation with polyhedral analysis to generate efficient matrix-free kernels and communication schedules for distributed block-structured PDE solvers.
-
Partitioning Unstructured Sparse Tensor Algebra for Load-Balanced Parallel Execution
A new partitioning algorithm that provably load-balances arbitrary sparse tensor algebra expressions by generalizing parallel merging to multi-operand, multi-dimensional hierarchical structures, implemented in a compiler framework.
-
ATLAAS: Automatic Tensor-Level Abstraction of Accelerator Semantics
ATLAAS automatically converts RTL-extracted bit-level accelerator semantics into tensor-level ISA specs via an 8-pass MLIR pipeline, enabling automated compiler backend generation for designs like Gemmini and VTA.
-
Decoupling Data Layouts from Bounding Volume Hierarchies
Scion is a new DSL and compiler that decouples BVH data layouts from traversal algorithms, enabling architecture-agnostic layout optimizations and a novel Pareto-optimal ray tracing layout.
-
NESA: Relational Neuro-Symbolic Static Program Analysis
NESA presents a neuro-symbolic framework that decomposes static analyses into policy-defined sub-problems solved by parsers and LLMs to enable compilation-free customizable analysis with reduced hallucinations.
-
The EDGE Language: Extended General Einsums for Graph Algorithms
EDGE extends Einsum notation with graph-specific operations to create a unified tensor-algebra framework for expressing and manipulating graph algorithms.