Irregular applications comprise an increasingly important workload domain for many fields, including bioinformatics, chemistry, physics, social sciences and machine learning. Therefore, achieving high performance and energy efficiency in the execution of emerging irregular applications is of vital importance. This dissertation studies the root causes of inefficiency of irregular applications in modern computing systems, and fundamentally addresses such inefficiencies, by proposing low-overhead synchronization techniques among parallel threads in cooperation with well-crafted data access policies. We make four major contributions to accelerating irregular applications in different contexts including CPU and Near-Data-Processing (NDP) (or Processing-In-Memory (PIM)) systems. First, we design ColorTM, a novel parallel graph coloring algorithm for CPU systems that trades off using synchronization with lower data access costs. Second, we propose SmartPQ, an adaptive priority queue that achieves high performance under all various contention scenarios in Non-Uniform Memory Access CPU systems. Third, we introduce SynCron, the first practical hardware synchronization mechanism tailored for NDP systems. Fourth, we design SparseP, the first library for high-performance Sparse Matrix Vector Multiplication on real PIM systems. We demonstrate that the execution of irregular applications in CPU and NDP/PIM architectures can be significantly accelerated by co-designing lightweight synchronization approaches along with well-crafted data access policies. This dissertation bridges the gap between processor-centric CPU systems and memory-centric PIM systems in the critically-important area of irregular applications. We hope that this dissertation inspires future work in co-designing software algorithms with cutting-edge computing platforms to significantly accelerate emerging irregular applications.
翻译:非常规应用包括许多领域越来越重要的工作量领域,包括生物信息学、化学、物理、社会科学和机器学习。因此,在新兴非常规应用的实施中实现高性能和能源效率至关重要。这一论文研究现代计算机系统中非正常应用效率低下的根源,并从根本上解决这种低效率问题,方法是与精心设计的数据访问政策合作,在平行线中提出低超性能同步技术,在平行线中提出低超性能同步技术;我们为加快不同背景下的不规则应用做出了四大贡献,包括:CPU 和 Near-Data-Process(NDP) (或处理-内装(PIM)系统)系统。首先,我们设计了ColoremTM,这是使用同步和较低的数据访问成本进行交易的CPU系统新的平行图形颜色算法。我们提议SmartPQ,一个适应性优先排队,在非统一记忆存取数据访问的各类争议情景下取得高性成绩。第三,我们引入SynCron,这是为NDP系统定制的第一个实用的硬件同步应用系统。第四,我们设计了Sprass-crealPla Plapper-cal Pold-cilation lical-de-deal-deal-deal limalation lider-deplate listration listrational lipal lishing lishmental lishing lishing astepal lipal lishmental lishmentaldalking lishal listring asteutmental lishalkingsteutmental listeutdal lishmental lishmental lishmental lishmental listeftal str astedal listr str asteal liftal str lical lical lical lical lical listral lical licaldal lial stral listedal lial lical stral stral lial stral listal listal lical listal listaldal stral listal stral listedal