\relax
\providecommand\hyper@newdestlabel[2]{}
\providecommand\HyField@AuxAddToFields[1]{}
\providecommand\HyField@AuxAddToCoFields[2]{}
\bibstyle{IEEEtran}
\citation{10.1093/comjnl/5.1.10}
\citation{frigoCacheobliviousAlgorithms1999b,wulfHittingMemoryWall1995}
\citation{cormenIntroductionAlgorithms2022}
\citation{kaligosiHowBranchMispredictions2006}
\citation{axtmannInPlaceParallelSuper2017}
\@writefile{toc}{\contentsline {section}{\numberline {I}Introduction}{1}{section.1}\protected@file@percent }
\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {I-A}}Motivation: From Comparison Trees to Spatial Classification}{1}{subsection.1.1}\protected@file@percent }
\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {I-B}}Spatial Mapping and Hardware-Conscious Processing}{1}{subsection.1.2}\protected@file@percent }
\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {I-C}}Scientific Contributions}{1}{subsection.1.3}\protected@file@percent }
\citation{blellochPrefixSumsTheir2018}
\citation{blumofeSchedulingMultithreadedComputations1999}
\citation{devroyeNonUniformRandomVariate1986}
\@writefile{lof}{\contentsline {figure}{\numberline {1}{\ignorespaces The "Spatial Advantage": ARS Aero Speedup across diverse data domains ($N=10^7$) on i64 datatype and random strings. The framework demonstrates significant gains in high-entropy and complex-type sorting while maintaining competitive parity on low-entropy duplicates.}}{2}{figure.caption.1}\protected@file@percent }
\providecommand*\caption@xref[2]{\@setref\relax\@undefined{#1}}
\newlabel{fig:teaser}{{1}{2}{The "Spatial Advantage": ARS Aero Speedup across diverse data domains ($N=10^7$) on i64 datatype and random strings. The framework demonstrates significant gains in high-entropy and complex-type sorting while maintaining competitive parity on low-entropy duplicates}{figure.caption.1}{}}
\newlabel{fig:teaser@cref}{{[figure][1][]1}{[1][1][]2}{}{}{}}
\@writefile{toc}{\contentsline {section}{\numberline {II}The ARS Evolutionary Phylogeny}{2}{section.2}\protected@file@percent }
\citation{graefeImplementingSortingDatabase2006}
\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {II-A}}Exploratory Architectures}{3}{subsection.2.1}\protected@file@percent }
\@writefile{lof}{\contentsline {figure}{\numberline {2}{\ignorespaces Evolution of ARS execution time against PDQsort (i64 Random).}}{3}{figure.caption.2}\protected@file@percent }
\newlabel{fig:evolution}{{2}{3}{Evolution of ARS execution time against PDQsort (i64 Random)}{figure.caption.2}{}}
\newlabel{fig:evolution@cref}{{[figure][2][]2}{[1][3][]3}{}{}{}}
\@writefile{toc}{\contentsline {section}{\numberline {III}Architectural Design}{3}{section.3}\protected@file@percent }
\newlabel{sec:methodology}{{III}{3}{Architectural Design}{section.3}{}}
\newlabel{sec:methodology@cref}{{[section][3][]III}{[1][3][]3}{}{}{}}
\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {III-A}}Spatial Classification Pipeline}{3}{subsection.3.1}\protected@file@percent }
\citation{nybergAlphasortCachesensitiveParallel1995}
\@writefile{lof}{\contentsline {figure}{\numberline {3}{\ignorespaces The ARS execution pipeline. The framework transforms sorting into a hardware-conscious spatial classification pipeline built around parallel partitioning and cache-efficient memory movement.}}{4}{figure.caption.3}\protected@file@percent }
\newlabel{fig:pipeline}{{3}{4}{The ARS execution pipeline. The framework transforms sorting into a hardware-conscious spatial classification pipeline built around parallel partitioning and cache-efficient memory movement}{figure.caption.3}{}}
\newlabel{fig:pipeline@cref}{{[figure][3][]3}{[1][3][]4}{}{}{}}
\newlabel{eq:prefix_sum}{{1}{4}{Spatial Classification Pipeline}{equation.1}{}}
\newlabel{eq:prefix_sum@cref}{{[equation][1][]1}{[1][4][]4}{}{}{}}
\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {III-B}}Aero: Quantile-Adaptive Classification}{4}{subsection.3.2}\protected@file@percent }
\citation{zahariaDiscretizedStreamsFaulttolerant2013}
\citation{oneilLogstructuredMergetreeLSMtree1996}
\citation{devroyeNonUniformRandomVariate1986}
\@writefile{loa}{\contentsline {algorithm}{\numberline {1}{\ignorespaces Aero Division-Free Quantile Mapping}}{5}{algorithm.1}\protected@file@percent }
\newlabel{alg:aero_mapping}{{1}{5}{Aero Division-Free Quantile Mapping}{algorithm.1}{}}
\newlabel{alg:aero_mapping@cref}{{[algorithm][1][]1}{[1][4][]5}{}{}{}}
\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {III-C}}Exp D: Streaming Spatial Consolidation}{5}{subsection.3.3}\protected@file@percent }
\@writefile{loa}{\contentsline {algorithm}{\numberline {2}{\ignorespaces Concurrent Tiered Spatial Consolidation}}{5}{algorithm.2}\protected@file@percent }
\newlabel{alg:expd_merge}{{2}{5}{Concurrent Tiered Spatial Consolidation}{algorithm.2}{}}
\newlabel{alg:expd_merge@cref}{{[algorithm][2][]2}{[1][5][]5}{}{}{}}
\@writefile{toc}{\contentsline {section}{\numberline {IV}Theoretical Analysis}{5}{section.4}\protected@file@percent }
\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {IV-A}}Mathematical Formalization}{5}{subsection.4.1}\protected@file@percent }
\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {IV-B}}Inverse-CDF Approximation Invariants}{5}{subsection.4.2}\protected@file@percent }
\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {IV-C}}Classification and I/O Complexity}{6}{subsection.4.3}\protected@file@percent }
\@writefile{toc}{\contentsline {subsubsection}{\numberline {\mbox {IV-C}1}I/O Model and Cache Complexity}{6}{subsubsection.4.3.1}\protected@file@percent }
\@writefile{toc}{\contentsline {subsubsection}{\numberline {\mbox {IV-C}2}The "Memory Wall" Trade-off}{6}{subsubsection.4.3.2}\protected@file@percent }
\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {IV-D}}Complexity Analysis of Spatial Partitioning}{6}{subsection.4.4}\protected@file@percent }
\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {IV-E}}Space Complexity}{6}{subsection.4.5}\protected@file@percent }
\@writefile{toc}{\contentsline {section}{\numberline {V}Experimental Evaluation}{6}{section.5}\protected@file@percent }
\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {V-A}}Evaluation Goals}{6}{subsection.5.1}\protected@file@percent }
\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {V-B}}Experimental Environment}{6}{subsection.5.2}\protected@file@percent }
\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {V-C}}Benchmark Methodology}{7}{subsection.5.3}\protected@file@percent }
\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {V-D}}Throughput Scaling}{7}{subsection.5.4}\protected@file@percent }
\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {V-E}}Distribution Robustness}{7}{subsection.5.5}\protected@file@percent }
\newlabel{fig:sub_dist}{{4a}{8}{Distribution performance ($N=10^7$)}{figure.caption.4}{}}
\newlabel{fig:sub_dist@cref}{{[subfigure][1][4]4a}{[1][7][]8}{}{}{}}
\newlabel{sub@fig:sub_dist}{{a}{8}{Distribution performance ($N=10^7$)}{figure.caption.4}{}}
\newlabel{sub@fig:sub_dist@cref}{{[subfigure][1][4]4a}{[1][7][]8}{}{}{}}
\newlabel{fig:sub_entropy}{{4b}{8}{Entropy response ($N=10^7$)}{figure.caption.4}{}}
\newlabel{fig:sub_entropy@cref}{{[subfigure][2][4]4b}{[1][7][]8}{}{}{}}
\newlabel{sub@fig:sub_entropy}{{b}{8}{Entropy response ($N=10^7$)}{figure.caption.4}{}}
\newlabel{sub@fig:sub_entropy@cref}{{[subfigure][2][4]4b}{[1][7][]8}{}{}{}}
\@writefile{lof}{\contentsline {figure}{\numberline {4}{\ignorespaces Distribution and Entropy Performance Evaluation. ARS Aero maintains competitive performance across uniform and skewed distributions, while standard comparison algorithms exploit low-entropy duplicate patterns.}}{8}{figure.caption.4}\protected@file@percent }
\newlabel{fig:combined_performance}{{4}{8}{Distribution and Entropy Performance Evaluation. ARS Aero maintains competitive performance across uniform and skewed distributions, while standard comparison algorithms exploit low-entropy duplicate patterns}{figure.caption.4}{}}
\newlabel{fig:combined_performance@cref}{{[figure][4][]4}{[1][7][]8}{}{}{}}
\newlabel{fig:sub_scalability}{{5a}{8}{Parallel Scalability ($N=10^7$)}{figure.caption.5}{}}
\newlabel{fig:sub_scalability@cref}{{[subfigure][1][5]5a}{[1][7][]8}{}{}{}}
\newlabel{sub@fig:sub_scalability}{{a}{8}{Parallel Scalability ($N=10^7$)}{figure.caption.5}{}}
\newlabel{sub@fig:sub_scalability@cref}{{[subfigure][1][5]5a}{[1][7][]8}{}{}{}}
\newlabel{fig:sub_strings}{{5b}{8}{Sorting 10,000,000 Random Strings}{figure.caption.5}{}}
\newlabel{fig:sub_strings@cref}{{[subfigure][2][5]5b}{[1][7][]8}{}{}{}}
\newlabel{sub@fig:sub_strings}{{b}{8}{Sorting 10,000,000 Random Strings}{figure.caption.5}{}}
\newlabel{sub@fig:sub_strings@cref}{{[subfigure][2][5]5b}{[1][7][]8}{}{}{}}
\@writefile{lof}{\contentsline {figure}{\numberline {5}{\ignorespaces Scaling Performance under Parallel Cores and Type Complexity. ARS Aero maintains scaling advantages across multi-threaded integer and complex-type string workloads.}}{8}{figure.caption.5}\protected@file@percent }
\newlabel{fig:scaling_complex}{{5}{8}{Scaling Performance under Parallel Cores and Type Complexity. ARS Aero maintains scaling advantages across multi-threaded integer and complex-type string workloads}{figure.caption.5}{}}
\newlabel{fig:scaling_complex@cref}{{[figure][5][]5}{[1][7][]8}{}{}{}}
\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {V-F}}Robustness Analysis}{8}{subsection.5.6}\protected@file@percent }
\@writefile{lot}{\contentsline {table}{\numberline {I}{\ignorespaces Latency Stability ($N=10^7$, Median Time ms)}}{8}{table.caption.7}\protected@file@percent }
\newlabel{tab:latency_stability}{{I}{8}{Latency Stability ($N=10^7$, Median Time ms)}{table.caption.7}{}}
\newlabel{tab:latency_stability@cref}{{[table][1][]I}{[1][8][]8}{}{}{}}
\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {V-G}}Microarchitectural Analysis}{8}{subsection.5.7}\protected@file@percent }
\newlabel{fig:sub_constant_factor}{{6a}{9}{Effective Constant Factor ($T/N$)}{figure.caption.6}{}}
\newlabel{fig:sub_constant_factor@cref}{{[subfigure][1][6]6a}{[1][8][]9}{}{}{}}
\newlabel{sub@fig:sub_constant_factor}{{a}{9}{Effective Constant Factor ($T/N$)}{figure.caption.6}{}}
\newlabel{sub@fig:sub_constant_factor@cref}{{[subfigure][1][6]6a}{[1][8][]9}{}{}{}}
\newlabel{fig:sub_hardware_trends}{{6b}{9}{Hardware Utilization Trends}{figure.caption.6}{}}
\newlabel{fig:sub_hardware_trends@cref}{{[subfigure][2][6]6b}{[1][8][]9}{}{}{}}
\newlabel{sub@fig:sub_hardware_trends}{{b}{9}{Hardware Utilization Trends}{figure.caption.6}{}}
\newlabel{sub@fig:sub_hardware_trends@cref}{{[subfigure][2][6]6b}{[1][8][]9}{}{}{}}
\@writefile{lof}{\contentsline {figure}{\numberline {6}{\ignorespaces Algorithmic Complexity and Hardware Utilization Scaling. As dataset size $N$ increases, ARS constant factor stabilizes due to cache-conscious layout optimization despite scaling LLC miss rates.}}{9}{figure.caption.6}\protected@file@percent }
\newlabel{fig:combined_hardware_scaling}{{6}{9}{Algorithmic Complexity and Hardware Utilization Scaling. As dataset size $N$ increases, ARS constant factor stabilizes due to cache-conscious layout optimization despite scaling LLC miss rates}{figure.caption.6}{}}
\newlabel{fig:combined_hardware_scaling@cref}{{[figure][6][]6}{[1][8][]9}{}{}{}}
\@writefile{lof}{\contentsline {figure}{\numberline {7}{\ignorespaces Hardware Efficiency Comparison. ARS Aero achieves faster completion times despite a lower IPC due to a massive reduction in total instructions executed (bypassing comparison-tree branch overhead).}}{9}{figure.caption.8}\protected@file@percent }
\newlabel{fig:hardware_efficiency}{{7}{9}{Hardware Efficiency Comparison. ARS Aero achieves faster completion times despite a lower IPC due to a massive reduction in total instructions executed (bypassing comparison-tree branch overhead)}{figure.caption.8}{}}
\newlabel{fig:hardware_efficiency@cref}{{[figure][7][]7}{[1][9][]9}{}{}{}}
\@writefile{lot}{\contentsline {table}{\numberline {II}{\ignorespaces Throughput Scaling and Cache Pressure (Random i64)}}{9}{table.caption.9}\protected@file@percent }
\newlabel{tab:throughput}{{II}{9}{Throughput Scaling and Cache Pressure (Random i64)}{table.caption.9}{}}
\newlabel{tab:throughput@cref}{{[table][2][]II}{[1][9][]9}{}{}{}}
\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {V-H}}Streaming Evaluation}{9}{subsection.5.8}\protected@file@percent }
\@writefile{lot}{\contentsline {table}{\numberline {III}{\ignorespaces Exp D Streaming Performance Audit ($N=10^8$)}}{10}{table.caption.10}\protected@file@percent }
\newlabel{tab:streaming_audit}{{III}{10}{Exp D Streaming Performance Audit ($N=10^8$)}{table.caption.10}{}}
\newlabel{tab:streaming_audit@cref}{{[table][3][]III}{[1][9][]10}{}{}{}}
\@writefile{lof}{\contentsline {figure}{\numberline {8}{\ignorespaces Streaming Pareto Frontier ($N=10^8$). Performance varies based on data pattern; bursty workloads allow the shadow pipeline to achieve maximum efficiency.}}{10}{figure.caption.11}\protected@file@percent }
\newlabel{fig:pareto}{{8}{10}{Streaming Pareto Frontier ($N=10^8$). Performance varies based on data pattern; bursty workloads allow the shadow pipeline to achieve maximum efficiency}{figure.caption.11}{}}
\newlabel{fig:pareto@cref}{{[figure][8][]8}{[1][10][]10}{}{}{}}
\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {V-I}}Limitations and Trade-offs}{10}{subsection.5.9}\protected@file@percent }
\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {V-J}}Comparative Discussion}{10}{subsection.5.10}\protected@file@percent }
\@writefile{toc}{\contentsline {section}{\numberline {VI}Limitations}{10}{section.6}\protected@file@percent }
\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {VI-A}}Memory Overhead}{10}{subsection.6.1}\protected@file@percent }
\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {VI-B}}Duplicate-Dominated Workloads}{11}{subsection.6.2}\protected@file@percent }
\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {VI-C}}Cache-Capacity Dependence}{11}{subsection.6.3}\protected@file@percent }
\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {VI-D}}Quantile Estimation Accuracy}{11}{subsection.6.4}\protected@file@percent }
\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {VI-E}}Scope of Evaluation}{11}{subsection.6.5}\protected@file@percent }
\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {VI-F}}Streaming Scope}{11}{subsection.6.6}\protected@file@percent }
\@writefile{toc}{\contentsline {section}{\numberline {VII}Future Directions}{11}{section.7}\protected@file@percent }
\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {VII-A}}Recursive Cache-Aware Refinement}{11}{subsection.7.1}\protected@file@percent }
\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {VII-B}}Hierarchical Memory Staging}{11}{subsection.7.2}\protected@file@percent }
\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {VII-C}}Entropy-Adaptive Execution}{11}{subsection.7.3}\protected@file@percent }
\citation{kraska2018case}
\citation{khalifa2020learnedsort}
\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {VII-D}}Distributed Spatial Classification}{12}{subsection.7.4}\protected@file@percent }
\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {VII-E}}Accelerator-Aware Implementations}{12}{subsection.7.5}\protected@file@percent }
\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {VII-F}}Persistent Streaming Architectures}{12}{subsection.7.6}\protected@file@percent }
\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {VII-G}}Learned Spatial Mapping Functions}{12}{subsection.7.7}\protected@file@percent }
\@writefile{toc}{\contentsline {section}{\numberline {VIII}Conclusion}{12}{section.8}\protected@file@percent }
\bibdata{references}
\bibcite{10.1093/comjnl/5.1.10}{1}
\bibcite{frigoCacheobliviousAlgorithms1999b}{2}
\bibcite{wulfHittingMemoryWall1995}{3}
\bibcite{cormenIntroductionAlgorithms2022}{4}
\bibcite{kaligosiHowBranchMispredictions2006}{5}
\bibcite{axtmannInPlaceParallelSuper2017}{6}
\bibcite{blellochPrefixSumsTheir2018}{7}
\bibcite{blumofeSchedulingMultithreadedComputations1999}{8}
\bibcite{devroyeNonUniformRandomVariate1986}{9}
\bibcite{graefeImplementingSortingDatabase2006}{10}
\bibcite{nybergAlphasortCachesensitiveParallel1995}{11}
\bibcite{zahariaDiscretizedStreamsFaulttolerant2013}{12}
\bibcite{oneilLogstructuredMergetreeLSMtree1996}{13}
\bibcite{kraska2018case}{14}
\bibcite{khalifa2020learnedsort}{15}
\@writefile{toc}{\contentsline {section}{References}{13}{section*.12}\protected@file@percent }
\gdef \@abspage@last{13}