Collaborative Research: SHF: Medium: Co-Optimizing Computation and Data Transformations for Sparse Tensors

协作研究：SHF：中：稀疏张量的协同优化计算和数据转换

• 批准号: 2106621
• 负责人: David Lowenthal
• 金额: $ 39.74万
• 依托单位: University of Arizona
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-01-01 至 2026-12-31
• 项目状态: 未结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2106621&HistoricalAwards=false
• 关键词: Collaborative; Research; SHF; Medium; Co

Sparse tensor computations are central to important applications including computer-assisted drug design, fraud detection, and national security. Timely execution of these applications improves user productivity and reduces the energy consumption associated with each execution. Sparse computations are characterized as having inputs where many or most values are zero. To avoid the inefficiency of storing and computing on zero-valued data, applications only store the nonzeros, with auxiliary data structures to recover their locations. As a result, sparse tensor computations exhibit unpredictable memory-access patterns that include indirection through the auxiliary data structures. Consequently, on today’s computer architectures, performance of sparse tensor computations is completely dominated by the movement of data, through the memory system and across nodes. Data movement is expensive both in terms of execution time and energy expenditure. Optimizing data movement of sparse tensor computations as high-performance architectures have become increasingly diverse — conventional parallel architectures, graphics processors used as parallel accelerators and complex memory systems — creates a performance and productivity challenge for software developers who end up writing low-level architecture-specific code for each platform. The proposed approach simultaneously optimizes how data is organized in memory, how the computation is structured to access the data in a way that reduces data movement, and how the computation and data movement make best use of features of the hardware architectures. Since the nonzero structure of the data is unknown until program execution, the approach also examines runtime information in its decisions. The resulting co-optimization strategy enables a cohesive approach for iteratively making scheduling and data representation transformation decisions for a wide range of sparse computations and incorporating runtime adaptations.This project is developing a programming framework that permits high-level specification of a sparse computation and optimizes it to reduce data movement. It composes data representations, data layouts and storage mappings, and parallel schedules for sparse computations. It employs data dependencies, runtime information, and architecture features to fully bind the final generated code. This approach is intended to enable handling sparse tensor computations with dependences such as sparse triangular solve and many other solvers for systems of linear equations, applying reorderings such as Morton ordering on sparse tensors, and late binding of sparse tensor data representations. The novel and most significant aspects of the research include: (1) composable schedule and data transformations, including data layout transformations and storage mapping; (2) inspector synthesis for runtime data transformations between data representations, layouts, and storage mappings, which are composed with external functions; (3) support for data-dependent tensor computations; and, (4) framework abstractions deployed in the MLIR/LLVM compiler.The researchers are strongly committed to broadening participation in computing and have comprehensive plans to engage the underrepresented groups.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

稀疏张量计算是重要应用的核心，包括计算机辅助药物设计、欺诈检测和国家安全。这些应用的及时执行可以提高用户的生产力并减少与每次执行相关的能耗。大多数值都是零，为了避免零值数据的存储和计算效率低下，应用程序仅存储非零值，并使用辅助数据结构来恢复它们的位置，从而导致稀疏张量计算。表现出不可预测的内存访问模式，包括通过所检查的辅助数据结构的间接访问，在当今的计算机体系结构上，稀疏张量计算的性能完全由数据移动决定，通过内存系统和跨节点数据移动都是昂贵的。随着高性能架构变得越来越多样化（传统的并行架构、用作并行加速器的图形处理器和复杂的内存系统），优化稀疏张量计算的数据移动可以提高性能。以及最终为每个平台编写低级架构特定代码的软件开发人员的生产力挑战。所提出的方法同时优化了数据在内存中的组织方式、计算的结构以减少数据移动的方式访问数据，以及计算和数据移动如何充分利用硬件架构的功能，因为数据的非零结构在程序执行之前是未知的，因此该方法还在其决策中检查运行时信息，从而实现了一种内聚的方法。用于迭代地制作用于各种稀疏计算的调度和数据表示转换决策并结合运行时适应。该项目正在开发一个编程框架，该框架允许稀疏计算的高级规范并对其进行优化以减少数据移动。它使用数据依赖性、运行时信息和架构功能来完全绑定最终生成的代码。这种方法旨在支持处理稀疏张量计算。具有稀疏三角求解器和线性方程组的许多其他求解器等依赖性，在稀疏张量上应用莫顿排序等重新排序，以及稀疏张量数据表示的后期绑定。该研究的新颖且最重要的方面包括：（1） (2) 检查器综合，用于数据表示、布局和存储映射之间的运行时数据转换，这些数据由外部组成函数；(3) 支持数据相关的张量计算；(4) MLIR/LLVM 编译器中部署的框架抽象。研究人员坚定地致力于扩大对计算的参与，并制定了全面的计划来吸引代表性不足的群体。通过使用基金会的智力价值和更广泛的影响审查标准进行评估，NSF 的法定使命被认为值得支持。

项目成果

Code Synthesis for Sparse Tensor Format Conversion and Optimization

稀疏张量格式转换和优化的代码综合

• DOI: 10.1145/3579990.3580021
• 发表时间: 2023-02
• 期刊: ACM
• 作者: Popoola, Tobi;Zhao, Tuowen;St. George, Aaron;Bhetwal, Kalyan;Strout, Michelle Mills;Hall, Mary;Olschanowsky, Catherine
• 通讯作者: Olschanowsky, Catherine

Polyhedral Specification and Code Generation of Sparse Tensor Contraction with Co-iteration

协同迭代稀疏张量收缩的多面体规范和代码生成

• DOI: 10.1145/3566054
• 发表时间: 2022-08-25
• 期刊: ACM Transactions on Architecture and Code Optimization
• 影响因子: 1.6
• 作者: Tuowen Zhao;Tobi Popoola;Mary W. Hall;C. Olschanowsky;M. Strout
• 通讯作者: M. Strout

Runtime Composition of Iterations for Fusing Loop-carried Sparse Dependence

用于融合循环携带稀疏依赖的迭代的运行时组合

• DOI: 10.1145/3581784.3607097
• 发表时间: 2023-11
• 期刊: ACM
• 作者: Cheshmi, Kazem;Strout, Michelle;Mehri Dehnavi, Maryam
• 通讯作者: Mehri Dehnavi, Maryam