diff --git a/exp/index_exp1_1.pdf b/exp/index_exp1_1.pdf index 43eddc2..8a90ec9 100644 Binary files a/exp/index_exp1_1.pdf and b/exp/index_exp1_1.pdf differ diff --git a/exp/index_exp1_2.pdf b/exp/index_exp1_2.pdf index fdbf14d..dcabdd8 100644 Binary files a/exp/index_exp1_2.pdf and b/exp/index_exp1_2.pdf differ diff --git a/exp/index_exp2_1.pdf b/exp/index_exp2_1.pdf index 5f95fae..baa8451 100644 Binary files a/exp/index_exp2_1.pdf and b/exp/index_exp2_1.pdf differ diff --git a/exp/index_exp2_2.pdf b/exp/index_exp2_2.pdf index 46e181c..868daae 100644 Binary files a/exp/index_exp2_2.pdf and b/exp/index_exp2_2.pdf differ diff --git a/exp/index_exp2_3.pdf b/exp/index_exp2_3.pdf new file mode 100644 index 0000000..c9b2286 Binary files /dev/null and b/exp/index_exp2_3.pdf differ diff --git a/exp/index_exp4_1.pdf b/exp/index_exp4_1.pdf index eadd288..3b6aaa2 100644 Binary files a/exp/index_exp4_1.pdf and b/exp/index_exp4_1.pdf differ diff --git a/exp/index_exp4_2.pdf b/exp/index_exp4_2.pdf index 1339637..3d360c0 100644 Binary files a/exp/index_exp4_2.pdf and b/exp/index_exp4_2.pdf differ diff --git a/fig/overview.png b/fig/overview.png new file mode 100644 index 0000000..b622822 Binary files /dev/null and b/fig/overview.png differ diff --git a/references.bib b/references.bib index ea56307..e807be6 100644 --- a/references.bib +++ b/references.bib @@ -608,3 +608,21 @@ pages = {324--339}, year = {2024} } + +@inproceedings{Xie12supercomputer, + author = {Bing Xie and + Jeffrey S. Chase and + David Dillow and + Oleg Drokin and + Scott Klasky and + Sarp Oral and + Norbert Podhorszki}, + editor = {Jeffrey K. Hollingsworth}, + title = {Characterizing output bottlenecks in a supercomputer}, + booktitle = {{SC} Conference on High Performance Computing Networking, Storage + and Analysis, {SC} '12, Salt Lake City, UT, {USA} - November 11 - + 15, 2012}, + pages = {8}, + publisher = {{IEEE/ACM}}, + year = {2012} +} diff --git a/rs_retrieval.aux b/rs_retrieval.aux index 4646d05..e4efaa6 100644 --- a/rs_retrieval.aux +++ b/rs_retrieval.aux @@ -28,11 +28,11 @@ \citation{WangK16MVOCC} \citation{Hong25HDCC} \citation{Wu25OOCC} -\citation{Hong25HDCC} \@writefile{toc}{\contentsline {section}{\numberline {II}Related Work}{2}{}\protected@file@percent } \newlabel{sec:RW}{{II}{2}} \@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {II-A}}I/O-Efficient Spatio-Temporal Retrieval Processing}{2}{}\protected@file@percent } \@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {II-B}}Concurrency Control}{2}{}\protected@file@percent } +\citation{Hong25HDCC} \citation{Peng26IOsurvey} \citation{Chen21Tuning1} \citation{Bez20TuningLayer} @@ -47,126 +47,136 @@ \newlabel{eqn:pre_rs}{{1}{3}} \newlabel{eqn:pre_st_query}{{2}{3}} \newlabel{eqn:cost_total}{{3}{3}} +\@writefile{lof}{\contentsline {figure}{\numberline {1}{\ignorespaces The workflow for processing concurrent spatio-temporal range retrievals in the system}}{4}{}\protected@file@percent } +\newlabel{fig:overview}{{1}{4}} \newlabel{eqn_pre_objective}{{4}{4}} -\@writefile{toc}{\contentsline {section}{\numberline {IV}I/O-aware Indexing Structure}{4}{}\protected@file@percent } -\newlabel{sec:Index}{{IV}{4}} -\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {IV-A}}Index schema design}{4}{}\protected@file@percent } -\newlabel{eqn_pre_gridkey}{{5}{4}} -\@writefile{lof}{\contentsline {figure}{\numberline {1}{\ignorespaces Index schema design}}{5}{}\protected@file@percent } -\newlabel{fig:index}{{1}{5}} -\@writefile{lof}{\contentsline {figure}{\numberline {2}{\ignorespaces Retrieval-time Execution}}{5}{}\protected@file@percent } -\newlabel{fig_ST_Query}{{2}{5}} -\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {IV-B}}Retrieval-time Execution}{5}{}\protected@file@percent } -\newlabel{eqn_pre_lookup_return}{{6}{5}} -\citation{Hong25HDCC} +\@writefile{toc}{\contentsline {section}{\numberline {IV}System Overview}{4}{}\protected@file@percent } +\newlabel{sec:Overview}{{IV}{4}} +\@writefile{toc}{\contentsline {section}{\numberline {V}I/O-aware Indexing Structure}{4}{}\protected@file@percent } +\newlabel{sec:Index}{{V}{4}} +\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {V-A}}Index schema design}{4}{}\protected@file@percent } +\@writefile{lof}{\contentsline {figure}{\numberline {2}{\ignorespaces Index schema design}}{5}{}\protected@file@percent } +\newlabel{fig:index}{{2}{5}} +\@writefile{lof}{\contentsline {figure}{\numberline {3}{\ignorespaces Retrieval-time Execution}}{5}{}\protected@file@percent } +\newlabel{fig_ST_Query}{{3}{5}} +\newlabel{eqn_pre_gridkey}{{5}{5}} +\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {V-B}}Retrieval-time Execution}{6}{}\protected@file@percent } +\newlabel{eqn_pre_lookup_return}{{6}{6}} \newlabel{eqn_pre_spatial_query}{{7}{6}} -\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {IV-C}}Why I/O-aware}{6}{}\protected@file@percent } -\@writefile{toc}{\contentsline {section}{\numberline {V}Hybrid Concurrency-Aware I/O Coordination}{6}{}\protected@file@percent } -\newlabel{sec:CC}{{V}{6}} -\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {V-A}}Retrieval Admission and I/O Plan Generation}{6}{}\protected@file@percent } -\newlabel{eq:io_plan}{{8}{6}} -\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {V-B}}Contention Estimation and Path Selection}{6}{}\protected@file@percent } -\newlabel{eqn_tuning_table}{{9}{6}} -\@writefile{lof}{\contentsline {figure}{\numberline {3}{\ignorespaces Hybrid Concurrency-Aware I/O Coordination.}}{7}{}\protected@file@percent } -\newlabel{fig:cc}{{3}{7}} -\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {V-C}}Deterministic Coordinated and Non-deterministic Execution}{7}{}\protected@file@percent } -\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {V-D}}Optimistic Read Execution and Completion}{8}{}\protected@file@percent } -\@writefile{toc}{\contentsline {section}{\numberline {VI}I/O Stack Tuning}{8}{}\protected@file@percent } -\newlabel{sec:Tuning}{{VI}{8}} -\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {VI-A}}Formulation of Online I/O Tuning}{8}{}\protected@file@percent } -\newlabel{eqn_tuning_table}{{10}{8}} -\newlabel{eqn_tuning_table}{{11}{8}} -\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {VI-B}}Surrogate-Assisted GMAB for Online I/O Tuning}{8}{}\protected@file@percent } +\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {V-C}}Why I/O-aware}{6}{}\protected@file@percent } +\citation{Hong25HDCC} +\@writefile{toc}{\contentsline {section}{\numberline {VI}Hybrid Concurrency-Aware I/O Coordination}{7}{}\protected@file@percent } +\newlabel{sec:CC}{{VI}{7}} +\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {VI-A}}Retrieval Admission and I/O Plan Generation}{7}{}\protected@file@percent } +\newlabel{eq:io_plan}{{8}{7}} +\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {VI-B}}Contention Estimation and Path Selection}{7}{}\protected@file@percent } +\newlabel{eqn_tuning_table}{{9}{7}} +\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {VI-C}}Deterministic Coordinated and Non-deterministic Execution}{7}{}\protected@file@percent } +\@writefile{lof}{\contentsline {figure}{\numberline {4}{\ignorespaces Hybrid Concurrency-Aware I/O Coordination.}}{8}{}\protected@file@percent } +\newlabel{fig:cc}{{4}{8}} +\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {VI-D}}Optimistic Read Execution and Completion}{8}{}\protected@file@percent } +\@writefile{toc}{\contentsline {section}{\numberline {VII}I/O Stack Tuning}{8}{}\protected@file@percent } +\newlabel{sec:Tuning}{{VII}{8}} +\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {VII-A}}Formulation of Online I/O Tuning}{8}{}\protected@file@percent } +\newlabel{eqn_tuning_table}{{10}{9}} +\newlabel{eqn_tuning_table}{{11}{9}} +\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {VII-B}}Surrogate-Assisted GMAB for Online I/O Tuning}{9}{}\protected@file@percent } \@writefile{loa}{\contentsline {algocf}{\numberline {1}{\ignorespaces Surrogate-Assisted Genetic Multi-Armed Bandit (SA-GMAB)}}{9}{}\protected@file@percent } \newlabel{alg:sa-gmab}{{1}{9}} -\@writefile{toc}{\contentsline {section}{\numberline {VII}Performance Evaluation}{9}{}\protected@file@percent } -\newlabel{sec:EXP}{{VII}{9}} -\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {VII-A}}Experimental Setup}{9}{}\protected@file@percent } -\@writefile{toc}{\contentsline {subsubsection}{\numberline {\mbox {VII-A}1}Dataset}{9}{}\protected@file@percent } -\@writefile{toc}{\contentsline {subsubsection}{\numberline {\mbox {VII-A}2}Retrieval Workload}{9}{}\protected@file@percent } +\citation{Xie12supercomputer} \@writefile{lot}{\contentsline {table}{\numberline {I}{\ignorespaces Dataset Statistics}}{10}{}\protected@file@percent } \newlabel{table_dataset}{{I}{10}} +\@writefile{toc}{\contentsline {section}{\numberline {VIII}Performance Evaluation}{10}{}\protected@file@percent } +\newlabel{sec:EXP}{{VIII}{10}} +\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {VIII-A}}Experimental Setup}{10}{}\protected@file@percent } +\@writefile{toc}{\contentsline {subsubsection}{\numberline {\mbox {VIII-A}1}Dataset}{10}{}\protected@file@percent } +\@writefile{toc}{\contentsline {subsubsection}{\numberline {\mbox {VIII-A}2}Retrieval Workload}{10}{}\protected@file@percent } \@writefile{lot}{\contentsline {table}{\numberline {II}{\ignorespaces Cluster Configurations}}{10}{}\protected@file@percent } \newlabel{table_config}{{II}{10}} -\newlabel{sec_exp_env}{{\mbox {VII-A}3}{10}} -\@writefile{toc}{\contentsline {subsubsection}{\numberline {\mbox {VII-A}3}Experimental Environment}{10}{}\protected@file@percent } -\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {VII-B}}Evaluating the Data Indexing Structure}{10}{}\protected@file@percent } -\newlabel{fig:index_exp1_1}{{\mbox {VII-B}1}{10}} -\newlabel{fig:index_exp1_2}{{\mbox {VII-B}1}{10}} -\@writefile{lof}{\contentsline {figure}{\numberline {4}{\ignorespaces The efficiency of I/O selectivity}}{10}{}\protected@file@percent } -\@writefile{lof}{\contentsline {subfigure}{\numberline{(a)}{\ignorespaces {Query footprint ratios}}}{10}{}\protected@file@percent } -\@writefile{lof}{\contentsline {subfigure}{\numberline{(b)}{\ignorespaces {Query spatial extents}}}{10}{}\protected@file@percent } -\newlabel{fig:index_exp1}{{4}{10}} -\newlabel{fig:index_exp2_1}{{\mbox {VII-B}2}{10}} -\newlabel{fig:index_exp2_2}{{\mbox {VII-B}2}{10}} -\@writefile{lof}{\contentsline {figure}{\numberline {5}{\ignorespaces End-to-End retrieval latency and latency breakdown}}{10}{}\protected@file@percent } -\@writefile{lof}{\contentsline {subfigure}{\numberline{(a)}{\ignorespaces {Query footprint ratios}}}{10}{}\protected@file@percent } -\@writefile{lof}{\contentsline {subfigure}{\numberline{(b)}{\ignorespaces {Various baselines}}}{10}{}\protected@file@percent } -\newlabel{fig:index_exp2}{{5}{10}} -\newlabel{sec:Index_exp_1}{{\mbox {VII-B}1}{10}} -\@writefile{toc}{\contentsline {subsubsection}{\numberline {\mbox {VII-B}1}I/O Selectivity Analysis}{10}{}\protected@file@percent } -\newlabel{sec:Index_exp_2}{{\mbox {VII-B}2}{10}} -\@writefile{toc}{\contentsline {subsubsection}{\numberline {\mbox {VII-B}2}End-to-End Retrieval Latency}{10}{}\protected@file@percent } -\newlabel{fig:index_exp3_1}{{\mbox {VII-B}3}{11}} -\newlabel{fig:index_exp3_2}{{\mbox {VII-B}3}{11}} -\@writefile{lof}{\contentsline {figure}{\numberline {6}{\ignorespaces Ablation analysis}}{11}{}\protected@file@percent } -\@writefile{lof}{\contentsline {subfigure}{\numberline{(a)}{\ignorespaces {I/O reduction analysis}}}{11}{}\protected@file@percent } -\@writefile{lof}{\contentsline {subfigure}{\numberline{(b)}{\ignorespaces {Latency breakdown}}}{11}{}\protected@file@percent } -\newlabel{fig:index_exp3}{{6}{11}} -\@writefile{lof}{\contentsline {figure}{\numberline {7}{\ignorespaces Impact of grid resolution on query latency}}{11}{}\protected@file@percent } -\newlabel{fig:index_exp3_3}{{7}{11}} -\newlabel{sec:Index_exp_3}{{\mbox {VII-B}3}{11}} -\@writefile{toc}{\contentsline {subsubsection}{\numberline {\mbox {VII-B}3}Ablation Study}{11}{}\protected@file@percent } -\@writefile{toc}{\contentsline {subsubsection}{\numberline {\mbox {VII-B}4}Index Construction and Storage Overhead}{11}{}\protected@file@percent } -\newlabel{fig:index_exp4_2}{{\mbox {VII-B}4}{12}} -\newlabel{fig:index_exp4_1}{{\mbox {VII-B}4}{12}} -\@writefile{lof}{\contentsline {figure}{\numberline {8}{\ignorespaces Index construction and storage overhead}}{12}{}\protected@file@percent } +\newlabel{sec_exp_env}{{\mbox {VIII-A}3}{10}} +\@writefile{toc}{\contentsline {subsubsection}{\numberline {\mbox {VIII-A}3}Experimental Environment}{10}{}\protected@file@percent } +\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {VIII-B}}Evaluating the Data Indexing Structure}{10}{}\protected@file@percent } +\newlabel{fig:index_exp1_1}{{\mbox {VIII-B}1}{11}} +\newlabel{fig:index_exp1_2}{{\mbox {VIII-B}1}{11}} +\@writefile{lof}{\contentsline {figure}{\numberline {5}{\ignorespaces The efficiency of I/O selectivity}}{11}{}\protected@file@percent } +\@writefile{lof}{\contentsline {subfigure}{\numberline{(a)}{\ignorespaces {Query footprint ratios}}}{11}{}\protected@file@percent } +\@writefile{lof}{\contentsline {subfigure}{\numberline{(b)}{\ignorespaces {Query spatial extents}}}{11}{}\protected@file@percent } +\newlabel{fig:index_exp1}{{5}{11}} +\newlabel{fig:index_exp2_1}{{\mbox {VIII-B}2}{11}} +\newlabel{fig:index_exp2_2}{{\mbox {VIII-B}2}{11}} +\@writefile{lof}{\contentsline {figure}{\numberline {6}{\ignorespaces End-to-End retrieval latency}}{11}{}\protected@file@percent } +\@writefile{lof}{\contentsline {subfigure}{\numberline{(a)}{\ignorespaces {Query footprint ratios}}}{11}{}\protected@file@percent } +\@writefile{lof}{\contentsline {subfigure}{\numberline{(b)}{\ignorespaces {Query footprint ratios}}}{11}{}\protected@file@percent } +\newlabel{fig:index_exp2}{{6}{11}} +\newlabel{sec:Index_exp_1}{{\mbox {VIII-B}1}{11}} +\@writefile{toc}{\contentsline {subsubsection}{\numberline {\mbox {VIII-B}1}I/O Selectivity Analysis}{11}{}\protected@file@percent } +\newlabel{sec:Index_exp_2}{{\mbox {VIII-B}2}{11}} +\@writefile{toc}{\contentsline {subsubsection}{\numberline {\mbox {VIII-B}2}End-to-End Retrieval Latency}{11}{}\protected@file@percent } +\@writefile{lof}{\contentsline {figure}{\numberline {7}{\ignorespaces Latency breakdown}}{11}{}\protected@file@percent } +\newlabel{fig:index_exp2_3}{{7}{11}} +\newlabel{sec:Index_exp_3}{{\mbox {VIII-B}3}{11}} +\@writefile{toc}{\contentsline {subsubsection}{\numberline {\mbox {VIII-B}3}Ablation Study}{11}{}\protected@file@percent } +\newlabel{fig:index_exp3_1}{{\mbox {VIII-B}3}{12}} +\newlabel{fig:index_exp3_2}{{\mbox {VIII-B}3}{12}} +\@writefile{lof}{\contentsline {figure}{\numberline {8}{\ignorespaces Ablation analysis}}{12}{}\protected@file@percent } +\@writefile{lof}{\contentsline {subfigure}{\numberline{(a)}{\ignorespaces {I/O reduction analysis}}}{12}{}\protected@file@percent } +\@writefile{lof}{\contentsline {subfigure}{\numberline{(b)}{\ignorespaces {Latency breakdown}}}{12}{}\protected@file@percent } +\newlabel{fig:index_exp3}{{8}{12}} +\@writefile{lof}{\contentsline {figure}{\numberline {9}{\ignorespaces Impact of grid resolution on query latency}}{12}{}\protected@file@percent } +\newlabel{fig:index_exp3_3}{{9}{12}} +\newlabel{fig:index_exp4_2}{{\mbox {VIII-B}4}{12}} +\newlabel{fig:index_exp4_1}{{\mbox {VIII-B}4}{12}} +\@writefile{lof}{\contentsline {figure}{\numberline {10}{\ignorespaces Index construction and storage overhead}}{12}{}\protected@file@percent } \@writefile{lof}{\contentsline {subfigure}{\numberline{(a)}{\ignorespaces {Ingested images ($10^4$)}}}{12}{}\protected@file@percent } \@writefile{lof}{\contentsline {subfigure}{\numberline{(b)}{\ignorespaces {Various index types}}}{12}{}\protected@file@percent } -\newlabel{fig:index_exp4}{{8}{12}} -\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {VII-C}}Evaluating the Concurrency Control}{12}{}\protected@file@percent } -\@writefile{toc}{\contentsline {subsubsection}{\numberline {\mbox {VII-C}1}Concurrency Scalability}{12}{}\protected@file@percent } -\@writefile{toc}{\contentsline {subsubsection}{\numberline {\mbox {VII-C}2}Storage-Level Effects and Request Collapse}{12}{}\protected@file@percent } -\newlabel{fig:cc_exp1_3}{{9(a)}{13}} +\newlabel{fig:index_exp4}{{10}{12}} +\@writefile{toc}{\contentsline {subsubsection}{\numberline {\mbox {VIII-B}4}Index Construction and Storage Overhead}{12}{}\protected@file@percent } +\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {VIII-C}}Evaluating the Concurrency Control}{12}{}\protected@file@percent } +\newlabel{fig:cc_exp1_3}{{11(a)}{13}} \newlabel{sub@fig:cc_exp1_3}{{(a)}{13}} -\newlabel{fig:cc_exp1_2}{{9(b)}{13}} +\newlabel{fig:cc_exp1_2}{{11(b)}{13}} \newlabel{sub@fig:cc_exp1_2}{{(b)}{13}} -\newlabel{fig:cc_exp1_1}{{9(c)}{13}} +\newlabel{fig:cc_exp1_1}{{11(c)}{13}} \newlabel{sub@fig:cc_exp1_1}{{(c)}{13}} -\@writefile{lof}{\contentsline {figure}{\numberline {9}{\ignorespaces Concurrency scalability analysis under varying spatial overlap ratios ($\sigma $).}}{13}{}\protected@file@percent } +\@writefile{lof}{\contentsline {figure}{\numberline {11}{\ignorespaces Concurrency scalability analysis under varying spatial overlap ratios ($\sigma $).}}{13}{}\protected@file@percent } \@writefile{lof}{\contentsline {subfigure}{\numberline{(a)}{\ignorespaces {$\sigma =0.4$}}}{13}{}\protected@file@percent } \@writefile{lof}{\contentsline {subfigure}{\numberline{(b)}{\ignorespaces {$\sigma =0.6$}}}{13}{}\protected@file@percent } \@writefile{lof}{\contentsline {subfigure}{\numberline{(c)}{\ignorespaces {$\sigma =0.8$}}}{13}{}\protected@file@percent } -\newlabel{fig:cc_exp1}{{9}{13}} -\newlabel{fig:cc_exp3_1}{{\mbox {VII-C}2}{13}} -\newlabel{fig:cc_exp3_2}{{\mbox {VII-C}2}{13}} -\@writefile{lof}{\contentsline {figure}{\numberline {10}{\ignorespaces The data volume reduction and request collapse}}{13}{}\protected@file@percent } +\newlabel{fig:cc_exp1}{{11}{13}} +\@writefile{toc}{\contentsline {subsubsection}{\numberline {\mbox {VIII-C}1}Concurrency Scalability}{13}{}\protected@file@percent } +\newlabel{fig:cc_exp3_1}{{\mbox {VIII-C}2}{13}} +\newlabel{fig:cc_exp3_2}{{\mbox {VIII-C}2}{13}} +\@writefile{lof}{\contentsline {figure}{\numberline {12}{\ignorespaces The data volume reduction and request collapse}}{13}{}\protected@file@percent } \@writefile{lof}{\contentsline {subfigure}{\numberline{(a)}{\ignorespaces {The number of clients}}}{13}{}\protected@file@percent } \@writefile{lof}{\contentsline {subfigure}{\numberline{(b)}{\ignorespaces {The number of clients}}}{13}{}\protected@file@percent } -\newlabel{fig:cc_exp3}{{10}{13}} -\@writefile{lof}{\contentsline {figure}{\numberline {11}{\ignorespaces Mode Switching}}{13}{}\protected@file@percent } -\newlabel{fig:cc_exp4}{{11}{13}} -\newlabel{sec:ModeSwitch}{{\mbox {VII-C}3}{13}} -\@writefile{toc}{\contentsline {subsubsection}{\numberline {\mbox {VII-C}3}Deterministic and Non-Deterministic Modes}{13}{}\protected@file@percent } -\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {VII-D}}Evaluating the I/O Tuning}{13}{}\protected@file@percent } -\citation{Rajesh24TunIO} -\newlabel{fig:tune_exp1_1}{{\mbox {VII-D}1}{14}} -\newlabel{fig:tune_exp1_2}{{\mbox {VII-D}1}{14}} -\@writefile{lof}{\contentsline {figure}{\numberline {12}{\ignorespaces Efficiency analysis of the tuning framework.}}{14}{}\protected@file@percent } +\newlabel{fig:cc_exp3}{{12}{13}} +\@writefile{toc}{\contentsline {subsubsection}{\numberline {\mbox {VIII-C}2}Storage-Level Effects and Request Collapse}{13}{}\protected@file@percent } +\@writefile{lof}{\contentsline {figure}{\numberline {13}{\ignorespaces Mode Switching}}{14}{}\protected@file@percent } +\newlabel{fig:cc_exp4}{{13}{14}} +\newlabel{sec:ModeSwitch}{{\mbox {VIII-C}3}{14}} +\@writefile{toc}{\contentsline {subsubsection}{\numberline {\mbox {VIII-C}3}Deterministic and Non-Deterministic Modes}{14}{}\protected@file@percent } +\newlabel{fig:tune_exp1_1}{{\mbox {VIII-D}1}{14}} +\newlabel{fig:tune_exp1_2}{{\mbox {VIII-D}1}{14}} +\@writefile{lof}{\contentsline {figure}{\numberline {14}{\ignorespaces Efficiency analysis of the tuning framework.}}{14}{}\protected@file@percent } \@writefile{lof}{\contentsline {subfigure}{\numberline{(a)}{\ignorespaces {Tuning steps}}}{14}{}\protected@file@percent } \@writefile{lof}{\contentsline {subfigure}{\numberline{(b)}{\ignorespaces {Time (mins)}}}{14}{}\protected@file@percent } -\newlabel{fig:tune_exp1}{{12}{14}} -\@writefile{toc}{\contentsline {subsubsection}{\numberline {\mbox {VII-D}1}Convergence Speed and Tuning Cost}{14}{}\protected@file@percent } -\@writefile{lof}{\contentsline {figure}{\numberline {13}{\ignorespaces Mode Switching}}{14}{}\protected@file@percent } -\newlabel{fig:tune_exp3}{{13}{14}} -\newlabel{eq:roti}{{12}{14}} -\@writefile{toc}{\contentsline {subsubsection}{\numberline {\mbox {VII-D}2}Adaptation to Workload Shifts}{14}{}\protected@file@percent } +\newlabel{fig:tune_exp1}{{14}{14}} +\@writefile{toc}{\contentsline {subsection}{\numberline {\mbox {VIII-D}}Evaluating the I/O Tuning}{14}{}\protected@file@percent } +\@writefile{toc}{\contentsline {subsubsection}{\numberline {\mbox {VIII-D}1}Convergence Speed and Tuning Cost}{14}{}\protected@file@percent } +\citation{Rajesh24TunIO} \bibstyle{IEEEtran} \bibdata{IEEEabrv,references} \bibcite{Ma15RS_bigdata}{1} \bibcite{Haut21DDL_RS}{2} \bibcite{LEWIS17datacube}{3} \bibcite{Yan21RS_manage1}{4} +\@writefile{lof}{\contentsline {figure}{\numberline {15}{\ignorespaces Mode Switching}}{15}{}\protected@file@percent } +\newlabel{fig:tune_exp3}{{15}{15}} +\newlabel{eq:roti}{{12}{15}} +\@writefile{toc}{\contentsline {subsubsection}{\numberline {\mbox {VIII-D}2}Adaptation to Workload Shifts}{15}{}\protected@file@percent } +\@writefile{toc}{\contentsline {section}{\numberline {IX}Conclusions}{15}{}\protected@file@percent } +\newlabel{sec:Con}{{IX}{15}} +\@writefile{toc}{\contentsline {section}{References}{15}{}\protected@file@percent } \bibcite{liu24mstgi}{5} \bibcite{Strobl08PostGIS}{6} \bibcite{Simoes16PostGIST}{7} @@ -191,9 +201,7 @@ \bibcite{Chen21Tuning1}{26} \bibcite{Bez20TuningLayer}{27} \bibcite{Yang22end-IO}{28} -\@writefile{toc}{\contentsline {section}{\numberline {VIII}Conclusions}{15}{}\protected@file@percent } -\newlabel{sec:Con}{{VIII}{15}} -\@writefile{toc}{\contentsline {section}{References}{15}{}\protected@file@percent } \bibcite{Behzad13HDF5}{29} \bibcite{Wang26RethinkingTuning}{30} +\bibcite{Xie12supercomputer}{31} \gdef \@abspage@last{16} diff --git a/rs_retrieval.bbl b/rs_retrieval.bbl index 12953fa..8a55be5 100644 --- a/rs_retrieval.bbl +++ b/rs_retrieval.bbl @@ -200,4 +200,12 @@ W.~Zhang, H.~Wang, Z.~Shi, Y.~Wu, M.~Li, T.~Li, F.~Wang, and D.~Feng, graph query,'' \emph{{IEEE} Trans. Parallel Distrib. Syst.}, vol.~37, no.~3, pp. 633--650, 2026. +\bibitem{Xie12supercomputer} +B.~Xie, J.~S. Chase, D.~Dillow, O.~Drokin, S.~Klasky, S.~Oral, and + N.~Podhorszki, ``Characterizing output bottlenecks in a supercomputer,'' in + \emph{{SC} Conference on High Performance Computing Networking, Storage and + Analysis, {SC} '12, Salt Lake City, UT, {USA} - November 11 - 15, 2012}, + J.~K. Hollingsworth, Ed.\hskip 1em plus 0.5em minus 0.4em\relax {IEEE/ACM}, + 2012, p.~8. + \end{thebibliography} diff --git a/rs_retrieval.blg b/rs_retrieval.blg index fc6a7fb..059c913 100644 --- a/rs_retrieval.blg +++ b/rs_retrieval.blg @@ -20,45 +20,45 @@ Warning--empty author in LEWIS17datacube Warning--empty booktitle in Lim17OCC Done. -You've used 30 entries, +You've used 31 entries, 4087 wiz_defined-function locations, - 1814 strings with 30569 characters, -and the built_in function-call counts, 28689 in all, are: -= -- 2234 -> -- 867 -< -- 262 -+ -- 469 -- -- 157 -* -- 1426 -:= -- 4009 -add.period$ -- 71 -call.type$ -- 30 -change.case$ -- 42 + 1821 strings with 30947 characters, +and the built_in function-call counts, 29476 in all, are: += -- 2286 +> -- 900 +< -- 263 ++ -- 484 +- -- 165 +* -- 1465 +:= -- 4128 +add.period$ -- 74 +call.type$ -- 31 +change.case$ -- 45 chr.to.int$ -- 615 -cite$ -- 33 -duplicate$ -- 1983 -empty$ -- 2188 -format.name$ -- 188 -if$ -- 6739 +cite$ -- 34 +duplicate$ -- 2045 +empty$ -- 2263 +format.name$ -- 197 +if$ -- 6923 int.to.chr$ -- 0 -int.to.str$ -- 30 -missing$ -- 367 -newline$ -- 115 -num.names$ -- 37 -pop$ -- 868 +int.to.str$ -- 31 +missing$ -- 383 +newline$ -- 118 +num.names$ -- 40 +pop$ -- 910 preamble$ -- 1 purify$ -- 0 quote$ -- 2 -skip$ -- 2168 +skip$ -- 2224 stack$ -- 0 -substring$ -- 1482 -swap$ -- 1721 -text.length$ -- 57 +substring$ -- 1485 +swap$ -- 1763 +text.length$ -- 58 text.prefix$ -- 0 top$ -- 5 -type$ -- 30 +type$ -- 31 warning$ -- 3 -while$ -- 133 -width$ -- 32 -write$ -- 325 +while$ -- 136 +width$ -- 33 +write$ -- 335 (There were 3 warnings) diff --git a/rs_retrieval.log b/rs_retrieval.log index 1d76c63..e74ce5a 100644 --- a/rs_retrieval.log +++ b/rs_retrieval.log @@ -1,4 +1,4 @@ -This is pdfTeX, Version 3.141592653-2.6-1.40.25 (MiKTeX 23.4) (preloaded format=pdflatex 2025.10.23) 2 FEB 2026 20:11 +This is pdfTeX, Version 3.141592653-2.6-1.40.25 (MiKTeX 23.4) (preloaded format=pdflatex 2025.10.23) 3 FEB 2026 21:05 entering extended mode restricted \write18 enabled. %&-line parsing enabled. @@ -437,112 +437,132 @@ File: umsb.fd 2013/01/14 v3.01 AMS symbols B d}}{D:/software/ctex/MiKTeX/fonts/enc/dvips/base/8r.enc} -] -Underfull \hbox (badness 2318) in paragraph at lines 102--103 +] [2] +Underfull \hbox (badness 2318) in paragraph at lines 103--104 []\OT1/ptm/m/n/10 Overall, ex-ist-ing con-cur-rency con-trol mech-a-nisms are [] -Underfull \hbox (badness 1360) in paragraph at lines 102--103 +Underfull \hbox (badness 1360) in paragraph at lines 103--104 \OT1/ptm/m/n/10 largely de-signed around transaction-level cor-rect-ness and [] -[2] [3] -Overfull \hbox (17.09525pt too wide) detected at line 149 -[][] [] []\OML/cmm/m/it/10 ; - [] - - -Underfull \hbox (badness 2368) in paragraph at lines 153--154 +[3] +Underfull \hbox (badness 2368) in paragraph at lines 154--155 \OT1/ptm/m/n/10 band-width with-out caus-ing star-va-tion or ex-ces-sive [] - + +File: fig/overview.png Graphic file (type png) + +Package pdftex.def Info: fig/overview.png used on input line 160. +(pdftex.def) Requested size: 158.99377pt x 129.60448pt. + File: fig/index.png Graphic file (type png) -Package pdftex.def Info: fig/index.png used on input line 161. +Package pdftex.def Info: fig/index.png used on input line 174. (pdftex.def) Requested size: 464.39685pt x 100.8524pt. - + File: fig/st-query.png Graphic file (type png) -Package pdftex.def Info: fig/st-query.png used on input line 168. +Package pdftex.def Info: fig/st-query.png used on input line 181. (pdftex.def) Requested size: 158.99377pt x 365.21727pt. -[4] -Underfull \hbox (badness 2293) in paragraph at lines 211--214 +[4 <./fig/overview.png>] +Underfull \hbox (badness 2293) in paragraph at lines 224--227 []\OT1/ptm/m/n/10 The I/O-aware in-dex en-ables ef-fi-cient spatio-temporal [] -Underfull \hbox (badness 4752) in paragraph at lines 211--214 +Underfull \hbox (badness 4752) in paragraph at lines 224--227 \OT1/ptm/m/n/10 range re-trievals by di-rectly trans-lat-ing re-trieval pred-i- [] -Underfull \hbox (badness 3735) in paragraph at lines 211--214 +Underfull \hbox (badness 3735) in paragraph at lines 224--227 \OT1/ptm/m/n/10 cates into win-dowed read plans, while avoid-ing both [] -Underfull \hbox (badness 5756) in paragraph at lines 211--214 +Underfull \hbox (badness 5756) in paragraph at lines 224--227 \OT1/ptm/m/n/10 Given a user-specified spatio-temporal re-trieval $\OML/cmm/m/i t/10 q \OT1/cmr/m/n/10 = [] -[5 <./fig/index.png> <./fig/st-query.png>] - + +Underfull \vbox (badness 10000) has occurred while \output is active [] + + [5 <./fig/index.png> <./fig/st-query.png>] [6] + File: fig/cc.png Graphic file (type png) -Package pdftex.def Info: fig/cc.png used on input line 264. +Package pdftex.def Info: fig/cc.png used on input line 277. (pdftex.def) Requested size: 464.39685pt x 151.16782pt. - [6] -Underfull \vbox (badness 6575) has occurred while \output is active [] - - [7 <./fig/cc.png>] -Underfull \hbox (badness 4518) in paragraph at lines 351--351 + [7] [8 <./fig/cc.png>] +Underfull \hbox (badness 4518) in paragraph at lines 364--364 []\OT1/ptm/b/n/10 Algorithm 1: \OT1/ptm/m/n/10 Surrogate-Assisted Ge-netic Mult i- [] LaTeX Font Info: Trying to load font information for OT1+pcr on input line 3 -59. +72. (D:\software\ctex\MiKTeX\tex/latex/psnfss\ot1pcr.fd File: ot1pcr.fd 2001/06/04 font definitions for OT1/pcr. -) [8] -Underfull \vbox (badness 1097) has occurred while \output is active [] +) +Underfull \vbox (badness 2856) has occurred while \output is active [] - -Overfull \hbox (0.88164pt too wide) in paragraph at lines 433--433 + [9] +Overfull \hbox (0.88164pt too wide) in paragraph at lines 446--446 []|\OT1/ptm/b/n/8 Resolution| [] -Overfull \hbox (2.45601pt too wide) in paragraph at lines 431--445 +Overfull \hbox (2.45601pt too wide) in paragraph at lines 444--458 [][] [] -[9] - +[10] + File: exp/index_exp1_1.pdf Graphic file (type pdf) -Package pdftex.def Info: exp/index_exp1_1.pdf used on input line 510. -(pdftex.def) Requested size: 111.27748pt x 90.60472pt. - +Package pdftex.def Info: exp/index_exp1_1.pdf used on input line 525. +(pdftex.def) Requested size: 111.27748pt x 95.88266pt. + File: exp/index_exp1_2.pdf Graphic file (type pdf) -Package pdftex.def Info: exp/index_exp1_2.pdf used on input line 516. -(pdftex.def) Requested size: 111.27748pt x 89.72505pt. - +Package pdftex.def Info: exp/index_exp1_2.pdf used on input line 531. +(pdftex.def) Requested size: 111.27748pt x 95.44283pt. + File: exp/index_exp2_1.pdf Graphic file (type pdf) -Package pdftex.def Info: exp/index_exp2_1.pdf used on input line 532. -(pdftex.def) Requested size: 111.27748pt x 94.12335pt. - +Package pdftex.def Info: exp/index_exp2_1.pdf used on input line 547. +(pdftex.def) Requested size: 111.27748pt x 94.56317pt. + File: exp/index_exp2_2.pdf Graphic file (type pdf) -Package pdftex.def Info: exp/index_exp2_2.pdf used on input line 538. -(pdftex.def) Requested size: 111.27748pt x 86.20642pt. - [10 <./exp/index_exp1_1.pdf> <./exp/index_exp1_2.pdf +Package pdftex.def Info: exp/index_exp2_2.pdf used on input line 553. +(pdftex.def) Requested size: 111.27748pt x 94.56317pt. + +File: exp/index_exp2_3.pdf Graphic file (type pdf) + +Package pdftex.def Info: exp/index_exp2_3.pdf used on input line 561. +(pdftex.def) Requested size: 130.08621pt x 98.20985pt. + +File: exp/index_exp3_1.pdf Graphic file (type pdf) + +Package pdftex.def Info: exp/index_exp3_1.pdf used on input line 577. +(pdftex.def) Requested size: 105.4204pt x 80.83417pt. + +File: exp/index_exp3_2.pdf Graphic file (type pdf) + +Package pdftex.def Info: exp/index_exp3_2.pdf used on input line 583. +(pdftex.def) Requested size: 105.4204pt x 81.6675pt. + +File: exp/index_exp3_3.pdf Graphic file (type pdf) + +Package pdftex.def Info: exp/index_exp3_3.pdf used on input line 591. +(pdftex.def) Requested size: 130.08621pt x 104.8943pt. + [11 <./exp/index_exp1_1.pdf> <./exp/index_exp1_2.pdf pdfTeX warning: pdflatex.exe (file ./exp/index_exp1_2.pdf): PDF inclusion: mult iple pdfs with page group included in a single page @@ -554,33 +574,43 @@ iple pdfs with page group included in a single page pdfTeX warning: pdflatex.exe (file ./exp/index_exp2_2.pdf): PDF inclusion: mult iple pdfs with page group included in a single page +> <./exp/index_exp2_3.pdf + +pdfTeX warning: pdflatex.exe (file ./exp/index_exp2_3.pdf): PDF inclusion: mult +iple pdfs with page group included in a single page >] - -File: exp/index_exp3_1.pdf Graphic file (type pdf) - -Package pdftex.def Info: exp/index_exp3_1.pdf used on input line 555. -(pdftex.def) Requested size: 105.4204pt x 80.83417pt. - -File: exp/index_exp3_2.pdf Graphic file (type pdf) - -Package pdftex.def Info: exp/index_exp3_2.pdf used on input line 561. -(pdftex.def) Requested size: 105.4204pt x 81.6675pt. - -File: exp/index_exp3_3.pdf Graphic file (type pdf) - -Package pdftex.def Info: exp/index_exp3_3.pdf used on input line 569. -(pdftex.def) Requested size: 130.08621pt x 104.8943pt. - + File: exp/index_exp4_2.pdf Graphic file (type pdf) -Package pdftex.def Info: exp/index_exp4_2.pdf used on input line 587. +Package pdftex.def Info: exp/index_exp4_2.pdf used on input line 609. (pdftex.def) Requested size: 114.79138pt x 88.02173pt. - + File: exp/index_exp4_1.pdf Graphic file (type pdf) -Package pdftex.def Info: exp/index_exp4_1.pdf used on input line 593. +Package pdftex.def Info: exp/index_exp4_1.pdf used on input line 615. (pdftex.def) Requested size: 106.5929pt x 84.26234pt. - [11 <./exp/index_exp3_1.pdf> <./exp/index_exp3_2.pdf + +Underfull \hbox (badness 1939) in paragraph at lines 628--630 +[]\OT1/ptm/m/n/10 For com-par-i-son, we eval-u-ate the fol-low-ing ex-e-cu-tion + + [] + + +File: exp/cc_exp1_3.pdf Graphic file (type pdf) + +Package pdftex.def Info: exp/cc_exp1_3.pdf used on input line 637. +(pdftex.def) Requested size: 151.76744pt x 122.97867pt. + +File: exp/cc_exp1_2.pdf Graphic file (type pdf) + +Package pdftex.def Info: exp/cc_exp1_2.pdf used on input line 639. +(pdftex.def) Requested size: 151.76744pt x 122.97867pt. + +File: exp/cc_exp1_1.pdf Graphic file (type pdf) + +Package pdftex.def Info: exp/cc_exp1_1.pdf used on input line 643. +(pdftex.def) Requested size: 151.76744pt x 122.97867pt. +[12 <./exp/index_exp3_1.pdf> <./exp/index_exp3_2.pdf pdfTeX warning: pdflatex.exe (file ./exp/index_exp3_2.pdf): PDF inclusion: mult iple pdfs with page group included in a single page @@ -588,50 +618,25 @@ iple pdfs with page group included in a single page pdfTeX warning: pdflatex.exe (file ./exp/index_exp3_3.pdf): PDF inclusion: mult iple pdfs with page group included in a single page ->] -Underfull \hbox (badness 1939) in paragraph at lines 606--608 -[]\OT1/ptm/m/n/10 For com-par-i-son, we eval-u-ate the fol-low-ing ex-e-cu-tion +> <./exp/index_exp4_2.pdf - [] - - -File: exp/cc_exp1_3.pdf Graphic file (type pdf) - -Package pdftex.def Info: exp/cc_exp1_3.pdf used on input line 615. -(pdftex.def) Requested size: 151.76744pt x 122.97867pt. - -File: exp/cc_exp1_2.pdf Graphic file (type pdf) - -Package pdftex.def Info: exp/cc_exp1_2.pdf used on input line 617. -(pdftex.def) Requested size: 151.76744pt x 122.97867pt. - -File: exp/cc_exp1_1.pdf Graphic file (type pdf) - -Package pdftex.def Info: exp/cc_exp1_1.pdf used on input line 621. -(pdftex.def) Requested size: 151.76744pt x 122.97867pt. - -File: exp/cc_exp3_1.pdf Graphic file (type pdf) - -Package pdftex.def Info: exp/cc_exp3_1.pdf used on input line 645. -(pdftex.def) Requested size: 110.10678pt x 89.21667pt. - -File: exp/cc_exp3_2.pdf Graphic file (type pdf) - -Package pdftex.def Info: exp/cc_exp3_2.pdf used on input line 651. -(pdftex.def) Requested size: 113.62068pt x 88.02258pt. -[12 <./exp/index_exp4_2.pdf> <./exp/index_exp4_1.pdf +pdfTeX warning: pdflatex.exe (file ./exp/index_exp4_2.pdf): PDF inclusion: mult +iple pdfs with page group included in a single page +> <./exp/index_exp4_1.pdf pdfTeX warning: pdflatex.exe (file ./exp/index_exp4_1.pdf): PDF inclusion: mult iple pdfs with page group included in a single page >] - -File: exp/cc_exp4.pdf Graphic file (type pdf) - -Package pdftex.def Info: exp/cc_exp4.pdf used on input line 668. -(pdftex.def) Requested size: 130.08621pt x 107.97943pt. - -Underfull \vbox (badness 1708) has occurred while \output is active [] - + +File: exp/cc_exp3_1.pdf Graphic file (type pdf) + +Package pdftex.def Info: exp/cc_exp3_1.pdf used on input line 667. +(pdftex.def) Requested size: 110.10678pt x 89.21667pt. + +File: exp/cc_exp3_2.pdf Graphic file (type pdf) + +Package pdftex.def Info: exp/cc_exp3_2.pdf used on input line 673. +(pdftex.def) Requested size: 113.62068pt x 88.02258pt. [13 <./exp/cc_exp1_3.pdf> <./exp/cc_exp1_2.pdf pdfTeX warning: pdflatex.exe (file ./exp/cc_exp1_2.pdf): PDF inclusion: multipl @@ -648,66 +653,67 @@ e pdfs with page group included in a single page pdfTeX warning: pdflatex.exe (file ./exp/cc_exp3_2.pdf): PDF inclusion: multipl e pdfs with page group included in a single page -> <./exp/cc_exp4.pdf - -pdfTeX warning: pdflatex.exe (file ./exp/cc_exp4.pdf): PDF inclusion: multiple -pdfs with page group included in a single page >] -Underfull \hbox (badness 1874) in paragraph at lines 688--689 + +File: exp/cc_exp4.pdf Graphic file (type pdf) + +Package pdftex.def Info: exp/cc_exp4.pdf used on input line 690. +(pdftex.def) Requested size: 130.08621pt x 107.97943pt. + +Underfull \hbox (badness 1874) in paragraph at lines 710--711 \OT1/ptm/m/n/10 high-impact pa-ram-e-ter se-lec-tion and Re-in-force-ment [] - + File: exp/tune_exp1_1.pdf Graphic file (type pdf) -Package pdftex.def Info: exp/tune_exp1_1.pdf used on input line 699. +Package pdftex.def Info: exp/tune_exp1_1.pdf used on input line 721. (pdftex.def) Requested size: 110.10678pt x 90.52228pt. - + File: exp/tune_exp1_2.pdf Graphic file (type pdf) -Package pdftex.def Info: exp/tune_exp1_2.pdf used on input line 705. +Package pdftex.def Info: exp/tune_exp1_2.pdf used on input line 727. (pdftex.def) Requested size: 113.62068pt x 91.61536pt. - -File: exp/tune_exp3_1.pdf Graphic file (type pdf) - -Package pdftex.def Info: exp/tune_exp3_1.pdf used on input line 728. -(pdftex.def) Requested size: 130.08621pt x 111.06456pt. -[14 <./exp/tune_exp1_1.pdf> <./exp/tune_exp1_2.pdf +[14 <./exp/cc_exp4.pdf> <./exp/tune_exp1_1.pdf + +pdfTeX warning: pdflatex.exe (file ./exp/tune_exp1_1.pdf): PDF inclusion: multi +ple pdfs with page group included in a single page +> <./exp/tune_exp1_2.pdf pdfTeX warning: pdflatex.exe (file ./exp/tune_exp1_2.pdf): PDF inclusion: multi ple pdfs with page group included in a single page -> <./exp/tune_exp3_1.pdf - -pdfTeX warning: pdflatex.exe (file ./exp/tune_exp3_1.pdf): PDF inclusion: multi -ple pdfs with page group included in a single page >] -Underfull \hbox (badness 2495) in paragraph at lines 743--744 + +File: exp/tune_exp3_1.pdf Graphic file (type pdf) + +Package pdftex.def Info: exp/tune_exp3_1.pdf used on input line 750. +(pdftex.def) Requested size: 130.08621pt x 111.06456pt. + +Underfull \hbox (badness 2495) in paragraph at lines 765--766 []\OT1/ptm/m/n/10 This work is sup-ported by the Na-tional Key R&D [] -Underfull \hbox (badness 2799) in paragraph at lines 743--744 +Underfull \hbox (badness 2799) in paragraph at lines 765--766 \OT1/ptm/m/n/10 Pro-gram of China ``In-ter-gov-ern-men-tal In-ter-na-tional Sci - [] -Underfull \hbox (badness 7576) in paragraph at lines 743--744 +Underfull \hbox (badness 7576) in paragraph at lines 765--766 \OT1/ptm/m/n/10 ence and Tech-nol-ogy In-no-va-tion Co-op-er-a-tion" (Grant [] -(rs_retrieval.bbl [15]) [16 - -] (rs_retrieval.aux) +(rs_retrieval.bbl [15 <./exp/tune_exp3_1.pdf>]) [16] (rs_retrieval.aux) LaTeX Warning: There were multiply-defined labels. ) Here is how much of TeX's memory you used: - 5747 strings out of 476331 - 98061 string characters out of 5797649 + 5765 strings out of 476331 + 98457 string characters out of 5797649 1883660 words of memory out of 5000000 - 26053 multiletter control sequences out of 15000+600000 + 26069 multiletter control sequences out of 15000+600000 561830 words of font info for 131 fonts, out of 8000000 for 9000 1145 hyphenation exceptions out of 8191 62i,17n,67p,1675b,497s stack positions out of 10000i,1000n,20000p,200000b,200000s @@ -729,9 +735,9 @@ urier/ucrr8a.pfb> -Output written on rs_retrieval.pdf (16 pages, 2367217 bytes). +Output written on rs_retrieval.pdf (16 pages, 2406786 bytes). PDF statistics: - 404 PDF objects out of 1000 (max. 8388607) + 435 PDF objects out of 1000 (max. 8388607) 0 named destinations out of 1000 (max. 500000) - 106 words of extra memory for PDF output out of 10000 (max. 10000000) + 116 words of extra memory for PDF output out of 10000 (max. 10000000) diff --git a/rs_retrieval.pdf b/rs_retrieval.pdf index 406c5c0..1c4aaa9 100644 Binary files a/rs_retrieval.pdf and b/rs_retrieval.pdf differ diff --git a/rs_retrieval.synctex.gz b/rs_retrieval.synctex.gz index 7ad0446..f2381d7 100644 Binary files a/rs_retrieval.synctex.gz and b/rs_retrieval.synctex.gz differ diff --git a/rs_retrieval.tex b/rs_retrieval.tex index b4699ef..06f4e0e 100644 --- a/rs_retrieval.tex +++ b/rs_retrieval.tex @@ -81,11 +81,12 @@ To address the aforementioned problems, we propose a novel ``Index-as-an-Executi The remainder of this paper is organized as follows: Section~\ref{sec:RW} presents the related work. -Section~\ref{sec:DF} proposes the definition concerning the spatio-temporal range retrieval problem. -Section~\ref{sec:Index} proposes the indexing structure. -Section~\ref{sec:CC} proposes the hybrid concurrency control protocol. -Section~\ref{sec:Tuning} proposes the method of I/O stack tuning. -Section~\ref{sec:EXP} presents the experiments and results. +Section~\ref{sec:DF} formulates the spatio-temporal range retrieval problem and establishes the cost models. +Section~\ref{sec:Overview} provides an overview of the proposed framework and describes how the three modules are integrated. +Section~\ref{sec:Index} presents the I/O-aware indexing structure. +Section~\ref{sec:CC} proposes the hybrid concurrency-aware I/O coordination protocol. +Section~\ref{sec:Tuning} presents the GMAB-based online I/O stack tuning method. +Section~\ref{sec:EXP} presents the experiments and results. Section~\ref{sec:Con} concludes this paper with a summary. \section{Related Work}\label{sec:RW} @@ -145,7 +146,7 @@ Each retrieval $Q_i$ independently specifies a spatio-temporal window $\langle S \vspace{-0.05in} \begin{equation} \label{eqn_pre_objective} - \min \sum_{Q_i\in \mathcal{Q}}{\left( C_{meta}\left( Q_i \right) +\sum_{R\in \mathcal{R}_{Q_i}}{\left( C_{geo}\left( R,Q_i \right) +C_{io}\left( R,Q_i \right) \right)} \right)}, + \min \sum_{Q_i\in \mathcal{Q}} \bigl( C_{\text{meta}}(Q_i)+\sum_{R\in \mathcal{R}_{Q_i}} \bigl( C_{\text{geo}}(R,Q_i) + C_{\text{io}}(R,Q_i) \bigr) \bigr), \end{equation} subject to: \begin{enumerate} @@ -153,6 +154,18 @@ subject to: \item \textit{Isolation:} Concurrent reads must effectively share I/O bandwidth without causing starvation or excessive thrashing. \end{enumerate} +\section{System Overview}\label{sec:Overview} +\begin{figure} + \centering + \includegraphics[width=2.2in]{fig/overview.png} + \caption{The workflow for processing concurrent spatio-temporal range retrievals in the system} + \label{fig:overview} +\end{figure} + +To address the challenges of storage-level I/O contention and expensive runtime computations, we propose a layered distributed retrieval framework. As illustrated in Fig. \ref{fig:overview}, the system architecture is composed of four primary processing components: (1) \emph{requst interface}, (2) \emph{index manager}, (3) \emph{I/O coordinator}, (4) \emph{parallel executors}, and (5) \emph{adaptive tuner}. + +The $\emph{requst interface}$ serves as the system entry point. It is responsible for accepting concurrent spatio-temporal retrievals. The $\emph{index manager}$ acts as the planner of the system, interacting with the metadata storage. It translates logical spatio-temporal predicates into physical storage locations using a dual-layer inverted index. The $\emph{I/O coordinator}$ serves as the traffic control layer. It detects spatial overlaps among concurrent reading plans to identify potential I/O conflicts and applies the hybrid concurrency-aware protocol to reorder or merge conflicting requests. Finally, the $\emph{parallel executors}$ interface with the distributed file system or object store to read the pixel data. What's more, \emph{adaptive tuner} optimizes the execution parameters in the background. + \section{I/O-aware Indexing Structure}\label{sec:Index} This section introduces the details of the indexing structure for spatio-temporal range retrieval over RS data. @@ -452,6 +465,8 @@ To evaluate the system performance under diverse scenarios, we developed a synth \item Concurrency \& Contention: The number of concurrent clients $N$ varies from 1 to 64. To test the coordination mechanism, we control the Spatial Overlap Ratio $\sigma \in [0, 0.9]$ to simulate workloads ranging from disjoint access to highly concentrated hotspots. \end{itemize} +It is worth noting that, given the data-intensive nature of retrievals where a single request triggers GB-scale I/O and complex decoding, 64 concurrent streams are sufficient to fully saturate the aggregate I/O bandwidth and CPU resources of our experimental cluster. With 8 worker nodes connected via 10GbE, a concurrency of 64 implies an average of 8 heavy I/O threads per node. Previous characterization studies on Lustre-based supercomputers \cite{Xie12supercomputer} have revealed that client-side flow control typically limits in-flight RPCs to 8 concurrent requests and that exceeding this parallelism level exacerbates resource contention and straggler effects. Therefore, this setting represents a realistic heavy-load scenario where I/O interference significantly impacts performance. + \subsubsection{Experimental Environment} \label{sec_exp_env} All experiments are conducted on a cluster with 9 homogenous nodes (1 master node and 8 worker nodes). The cluster is connected via a 10Gbps high-speed Ethernet to ensure that network bandwidth is not the primary bottleneck compared to storage I/O. Table \ref{table_config} lists the detailed hardware and software configurations. The I/O-aware index (G2I/I2G) is deployed on HBase, while the raw image data is served by the Lustre parallel file system. @@ -531,16 +546,23 @@ First, we evaluated the effectiveness of data reduction by measuring the I/O sel \end{minipage} } \label{fig:index_exp2_1} - \subfigure[Various baselines]{ + \subfigure[Query footprint ratios]{ \begin{minipage}[b]{0.227\textwidth} \includegraphics[width=0.95\textwidth]{exp/index_exp2_2.pdf} \end{minipage} } \label{fig:index_exp2_2} - \caption{End-to-End retrieval latency and latency breakdown} + \caption{End-to-End retrieval latency} \label{fig:index_exp2} \end{figure} +\begin{figure} + \centering + \includegraphics[width=1.8in]{exp/index_exp2_3.pdf} + \caption{Latency breakdown} + \label{fig:index_exp2_3} +\end{figure} + We next measured the end-to-end retrieval latency to verify whether the I/O reduction translates into time efficiency. Fig.~\ref{fig:index_exp2}(a) reports the mean and 95th percentile (P95) latency across varying retrieval footprint ratios. The results reveal three distinct performance behaviors: Baseline 1 shows a high and flat latency curve ($\approx 4500$ ms), dominated by the cost of transferring entire images. Baseline 2, despite its optimal I/O selectivity, exhibits a significant latency floor ($\approx 380$ ms for small tile-level retrievals). This overhead stems from the on-the-fly geospatial computations required to calculate precise read windows. Ours achieves the lowest latency, ranging from 34 ms to 59 ms for typical tile-level retrievals. Crucially, for small-to-medium retrievals, our method outperforms Baseline 2 by an order of magnitude. The gap between the two curves highlights the advantage of our deterministic indexing approach: by pre-materializing grid-to-window mappings, we eliminate runtime coordinate transformations. Although our I/O volume is slightly larger (as shown in Sec.~\ref{sec:Index_exp_1}), the time saved by avoiding computational overhead far outweighs the cost of transferring a few extra kilobytes of padding data. To empirically validate the cost model proposed in Eq.~\ref{eqn:cost_total}, we further decomposed the retrieval latency into three components: metadata lookup ($C_{meta}$), geospatial computation ($C_{geo}$), and I/O access ($C_{io}$). Fig.~\ref{fig:index_exp2}(b) presents the time consumption breakdown for a representative medium-scale retrieval (involving approx. 50 image tiles). As expected, the latency of Baseline 1 is entirely dominated by $C_{io}$, rendering $C_{meta}$ and $C_{geo}$ negligible. The massive data transfer masks all other overheads. While $C_{io}$ of Baseline 2 is successfully reduced to the window size, a new bottleneck emerges in $C_{geo}$. The runtime coordinate transformations and polygon clipping consume nearly $40\%$ of the total execution time ($\approx 350 ms$). This observation confirms our theoretical analysis that window-based I/O shifts the bottleneck from storage to CPU. The proposed method exhibits a balanced profile. Although $C_{meta}$ increases slightly ($\approx 35 ms$) due to the two-phase index lookup (G2I + I2G), this cost is well-amortized. Crucially, $C_{geo}$ is effectively eliminated thanks to the pre-computed grid-window mappings. Consequently, our approach achieves a total latency of 580 ms, providing a $1.7\times$ speedup over Baseline 2 by removing the computational bottleneck without regressing on I/O performance. @@ -601,7 +623,7 @@ Finally, we evaluated the scalability and cost of maintaining the index. Fig.~\r \subsection{Evaluating the Concurrency Control} In this section, we evaluate the proposed hybrid coordination mechanism on a distributed storage cluster to assess its scalability, robustness under contention, and internal storage efficiency. -To systematically control the workload characteristics, we developed a synthetic workload generator. We define the Spatial Overlap Ratio ($\sigma$) to quantify the extent of shared data regions among concurrent queries, ranging from $\sigma=0$ (disjoint) to $\sigma=0.9$ (highly concentrated hotspots). The number of concurrent clients varies from $N=1$ to $N=64$. It is worth noting that, given the data-intensive nature of retrievals where a single request triggers GB-scale I/O and complex decoding, 64 concurrent streams are sufficient to fully saturate the aggregate I/O bandwidth and CPU resources of our experimental cluster, representing a heavy-load scenario in operational scientific computing environments. +To systematically control the workload characteristics, we developed a synthetic workload generator. We define the Spatial Overlap Ratio ($\sigma$) to quantify the extent of shared data regions among concurrent queries, ranging from $\sigma=0$ (disjoint) to $\sigma=0.9$ (highly concentrated hotspots). The number of concurrent clients varies from $N=1$ to $N=64$. For comparison, we evaluate the following execution schemes: \begin{enumerate}