TextGen：用于新型存储系统基准测试的真实文本数据集生成方法

Long-xiang WANG; Xiao-she DONG; Xing-jun ZHANG; Yin-feng WANG; Tao JU; Guo-fu FENG

doi:10.1631/FITEE.1500332

Your Location：

Home >

Browse articles >

TextGen：用于新型存储系统基准测试的真实文本数据集生成方法

常规文章 | Updated：2022-05-19

- TextGen：用于新型存储系统基准测试的真实文本数据集生成方法
- TextGen: a realistic text data content generation method for modern storage system benchmarks
- 信息与电子工程前沿（英文） 2016年17卷第10期页码：982-993
- Affiliations：
  
  1School of Electronic and Information Engineering, Xi'an Jiaotong University, Xi'an 710049, China
  2Shenzhen Institute of Information Technology, Shenzhen 518172, China
  3College of Information Technology, Shanghai Ocean University, Shanghai 201306, China
- Author bio：
  
  E-mail: wanglongxiang@stu.xjtu.edu.cn
  E-mail: xsdong@mail.xjtu.edu.cn
  E-mail: xjzhang@mail.xjtu.edu.cn
  E-mail: wangyinfeng@gmail.com
  E-mail: jutao2011@stu.xjtu.edu.cn
  E-mail: jt_f@163.com
- Funds：
  
  Project supported by the National Natural Science Foundation of China (Nos. 61572394 and 61272098), the Shenzhen Funda mental Research Plan (Nos. JCYJ20120615101127404 and JSGG20140519141854753), and the National Key Technologies R&D Program of China (No. 2011BAH04B03)
- DOI：10.1631/FITEE.1500332
  中图分类号： TP311.1
- 收稿：2015-10-13，
  
  录用：2016-3-21，
  
  纸质出版：2016-10
- Accepted：
Scan QR Code
TextGen：用于新型存储系统基准测试的真实文本数据集生成方法[J]. 信息与电子工程前沿（英文）, 2016,17(10):982-993.

Long-xiang WANG, Xiao-she DONG, Xing-jun ZHANG, et al. TextGen: a realistic text data content generation method for modern storage system benchmarks[J]. Frontiers of Information Technology & Electronic Engineering, 2016, 17(10): 982-993.
TextGen：用于新型存储系统基准测试的真实文本数据集生成方法[J]. 信息与电子工程前沿（英文）, 2016,17(10):982-993. DOI： 10.1631/FITEE.1500332.

Long-xiang WANG, Xiao-she DONG, Xing-jun ZHANG, et al. TextGen: a realistic text data content generation method for modern storage system benchmarks[J]. Frontiers of Information Technology & Electronic Engineering, 2016, 17(10): 982-993. DOI： 10.1631/FITEE.1500332.

摘要

新型存储系统通过内置数据压缩功能提高性能，并节省存储空间。因此，数据内容会显著影响存储系统基准测试结果。由于真实数据集规模庞大，难以复制到目标测试系统，并且大多数数据集由于隐私性无法进行共享。因此，基准测试程序需要人工生成测试数据集。为了保证测试结果的准确性，需要根据影响存储系统性能的真实数据集特征信息生成数据。现有方法SDGen在字节级别上分析真实数据集内容分布特征，并以此生成数据集，因此能够保证内置字节级压缩算法的存储系统测试结果准确。但是SDGen并未分析真实数据集的词级别内容分布特征，因此不能保证内置词级别压缩算法的存储系统测试结果准确，本文提出了一种基于Lognormal概率分布模型的文本数据集生成方法TextGen。该方法根据真实数据集的词切分结果建立语料库，分析语料库中词的分布特征，利用最大似然估计得到词分布的Lognormal模型参数，根据模型采用蒙特卡洛方法生成数据内容。该方法生成数据集所消耗的时间只与生成数据集规模相关，具有线性的时间复杂度

(

)。本文收集了四种数据集验证方法有效性，并通过一种典型的词级别压缩算法——ETDC（End-Tagged Dense Code）进行测试。实验结果表明：相比SDGen，TextGen生成文本数据集性能更高，并且，生成数据集用于压缩测试后与真实数据集的压缩速率、压缩率相似程度更高。

Abstract

Modern storage systems incorporate data compressors to improve their performance and capacity. As a result

data content can significantly influence the result of a storage system benchmark. Because real-world proprietary datasets are too large to be copied onto a test storage system

and most data cannot be shared due to privacy issues

a benchmark needs to generate data synthetically. To ensure that the result is accurate

it is necessary to generate data content based on the characterization of real-world data properties that influence the storage system performance during the execution of a benchmark. The existing approach

called SDGen

cannot guarantee that the benchmark result is accurate in storage systems that have built-in word-based compressors. The reason is that SDGen characterizes the properties that influence compression performance only at the byte level

and no properties are characterized at the word level. To address this problem

we present TextGen

a realistic text data content generation method for modern storage system benchmarks. TextGen builds the word corpus by segmenting real-world text datasets

and creates a word-frequency distribution by counting each word in the corpus. To improve data generation performance

the word-frequency distribution is fitted to a lognormal distribution by maximum likelihood estimation. The Monte Carlo approach is used to generate synthetic data. The running time of TextGen generation depends only on the expected data size

which means that the time complexity of TextGen is

(

). To evaluate TextGen

four real-world datasets were used to perform an experiment. The experimental results show that

compared with SDGen

the compression performance and compression ratio of the datasets generated by TextGen deviate less from real-world datasets when end-tagged dense code

a representative of word-based compressors

is evaluated.

关键词

Keywords

references

N Agrawal , , , W.J Bolosky , , , J.R Douceur , , , 等 . . A five-year study of file-system metadata . . ACM Trans. Stor. , , 2007 . . 3 ( ( 3 ): ): 9.1 - - 9.32 . . DOI: 10.1145/1288783.1288788 http://doi.org/10.1145/1288783.1288788 . .

N Agrawal , , , A.C Arpaci-Dusseau , , , R.H Arpaci-Dusseau . . Generating realistic impressions for file-system benchmarking . . ACM Trans. Stor. , , 2009 . . 5 ( ( 4 ): ): 16.1 - - 16.30 . . DOI: 10.1145/1629080.1629086 http://doi.org/10.1145/1629080.1629086 . .

E Anderson , , , M Kallahalla , , , M Uysal , , , 等 . . Buttress: a toolkit for flexible and high fidelity I/O benchmarking . . 2004 . . Proc. USENIX Conf. on File and Storage Technologies . . 4 . .

T.G Armstrong , , , V Ponnekanti , , , D Borthakur , , , 等 . . Linkbench: a database benchmark based on the Facebook social graph . . 2013 . . Proc. ACM SIGMOD Int. Conf. onManagement of Data . . 1185 - - 1196 . . DOI: 10.1145/2463676.2465296 http://doi.org/10.1145/2463676.2465296 . .

R Arnold , , , T Bell . . A corpus for the evaluation of lossless compression algorithms . . 1997 . . Data Compression Conf. . . 201 - - 210 . . DOI: 10.1109/DCC.1997.582019 http://doi.org/10.1109/DCC.1997.582019 . .

H Baayen . . Statistical-models for word-frequency distributions-a linguistic evaluation . . Comput. Human. , , 1992 . . 26 ( ( 5-6 ): ): 347 - - 363 . . DOI: 10.1007/Bf00136980 http://doi.org/10.1007/Bf00136980 . .

T Bck . . Evolutionary Algorithms in Theory and Practice , , 1996 . . : : Oxford, UK Oxford University Press , , 120 . .

J Bonwick , , , M Ahrens , , , V Henson , , , 等 . . The Zettabyte File System . . 2003 . . Technical Report, Sun Microsystems,Inc., Santa Clara, USA . . . .

G.E.P Box , , , M.E Muller . . A note on the generation of random normal deviates . . Ann.. Math. Statist. , , 1958 . . 29 ( ( 2 ): ): 610 - - 611 . . DOI: 10.1214/aoms/1177706645 http://doi.org/10.1214/aoms/1177706645 . .

N.R Brisaboa , , , E Iglesias , , , G Navarro , , , 等 . . An efficient compression code for text databases . . Adv. Inform. Retriev. , , 2003 . . 2633 468 - - 481 . . DOI: 10.1007/3-540-36618-0_33 http://doi.org/10.1007/3-540-36618-0_33 . .

N.R Brisaboa , , , A Faria , , , G Navarro , , , 等 . . Lightweight natural language text compression . . Inform. Retriev. , , 2007 . . 10 ( ( 1 ): ): 1 - - 33 . . DOI: 10.1007/s10791-006-9001-9 http://doi.org/10.1007/s10791-006-9001-9 . .

N.R Brisaboa , , , A Faria , , , G Navarro . . New adaptive compressors for natural language text . . Softw.-Pract. Exper. , , 2008 . . 38 ( ( 13 ): ): 1429 - - 1450 . . DOI: 10.1002/spe.882 http://doi.org/10.1002/spe.882 . .

N.R Brisaboa , , , A Faria , , , G Navarro , , , 等 . . Dynamic lightweight text compression . . ACM Trans. Inform. Syst. , , 2010 . . 28 ( ( 3 ): ): 1 - - 32 . . DOI: 10.1145/1777432.1777433 http://doi.org/10.1145/1777432.1777433 . .

C.M Chilan . . IOzone: an Open Source File SystemBenchmark Tool . . 2005 . . Technical Report, the National Center for Supercomputing Applications Hierarchical DataFormat Group, University of Illinois at Urbana-Champaign, Illinois . . . .

B.F Cooper , , , A Silberstein , , , E Tam , , , 等 . . Benchmarking cloud serving systems with YCSB . . 2010 . . Proc. ACM Sym0: on Cloud Computing . . 143 - - 154 . . DOI: 10.1145/1807128.1807152 http://doi.org/10.1145/1807128.1807152 . .

D.E Difallah , , , A Pavlo , , , C Curino , , , 等 . . OLTP-bench: an extensible testbed for benchmarking relational databases . . Proc. VLDB Endow. , , 2013 . . 7 ( ( 4 ): ): 277 - - 288 . . DOI: 10.14778/2732240.2732246 http://doi.org/10.14778/2732240.2732246 . .

I Drago , , , E Bocchi , , , M Mellia , , , 等 . . Benchmarking personal cloud storage . . 2013 . . Proc. Conf. on Int. Measurement . . 205 - - 212 . . DOI: 10.1145/2504730.2504762 http://doi.org/10.1145/2504730.2504762 . .

J Dvorsk , , , J Pokorn , , , V Snáel . . Word-based compression methods and indexing for text retrieval systems . . Adv. Database Inform. Syst. , , 1999 . . 1691 76 - - 84 . . DOI: 10.1007/3-540-48252-0_6 http://doi.org/10.1007/3-540-48252-0_6 . .

A Faria , , , N.R Brisaboa , , , G Navarro , , , 等 . . Word-based self-indexes for natural language text . . ACMTrans. Inform. Syst. , , 2012 . . 30 ( ( 1 ): ): 1 - - 34 . . DOI: 10.1145/2094072.2094073 http://doi.org/10.1145/2094072.2094073 . .

R Gracia-Tinedo , , , D Harnik , , , D Naor , , , 等 . . SDGen: mimicking datasets for content generation in storage benchmarks . . 2015 . . Proc. USENIX Conf. on File and StorageTechnologies . . 317 - - 330 . . . .

R.N Horspool , , , G.V Cormack . . Constructing wordbased text compression algorithms . . 1992 . . Data Compression Conf. . . 62 - - 71 . . DOI: 10.1109/DCC.1992.227475 http://doi.org/10.1109/DCC.1992.227475 . .

K Lang . . Newsweeder: learning to filter netnews . . 1995 . . Proc. Int. Conf. on Machine Learning . . 331 - - 339 . . . .

A Li , , , X Yang , , , S Kandula , , , 等 . . Cloudcmp: comparing public cloud providers . . 2010 . . Proc. ACMSIGCOMM Conf. on Internet Measurement . . 1 - - 14 . . DOI: 10.1145/1879141.1879143 http://doi.org/10.1145/1879141.1879143 . .

W.T Li . . Random texts exhibit Zipf-law-like wordfrequency distribution . . IEEE Trans. Inform. Theor. , , 1992 . . 38 ( ( 6 ): ): 1842 - - 1845 . . DOI: 10.1109/18.165464 http://doi.org/10.1109/18.165464 . .

A Moffat , , , J Zobel , , , N Sharman . . Text compression for dynamic document databases . . IEEE Trans. Knowl. Database Eng. , , 1997 . . 9 ( ( 2 ): ): 302 - - 313 . . DOI: 10.1109/69.591454 http://doi.org/10.1109/69.591454 . .

I.J Myung . . Tutorial on maximum likelihood estimation . . J. Math. Psychol. , , 2003 . . 47 ( ( 1 ): ): 90 - - 100 . . DOI: 10.1016/S0022-2496(02)00028-7 http://doi.org/10.1016/S0022-2496(02)00028-7 . .

D.M.W Powers . . Applications and explanations of Zipf's law . . 1998 . . Proc. Joint Conf. on New Methods in Language Processing and Computational NaturalLanguage Learning . . 151 - - 160 . . . .

O Rodeh , , , J Bacik , , , C Mason . . BTRFS: the LinuxB-tree filesystem . . ACM Trans. Stor. , , 2013 . . 9 ( ( 3 ): ): 1 - - 32 . . DOI: 10.1145/2501620.2501623 http://doi.org/10.1145/2501620.2501623 . .

D Salomon . . Data Compression: the Complete Reference , , 2006 . . : : New York,USA Springer-Verlag New York, Inc. , , 885 . .

V Tarasov , , , S Bhanage , , , E Zadok , , , 等 . . Benchmarking file system benchmarking: it *is* rocket science . . 2011 . . Proc. USENIX Conf. on Hot Topics in OperatingSystems . . 8 - - 13 . . . .

A Traeger , , , E Zadok , , , N Joukov , , , 等 . . A nine year study of file system and storage benchmarking . . ACM Trans. Stor. , , 2008 . . 4 ( ( 2 ): ): 1 - - 56 . . DOI: 10.1145/1367829.1367831 http://doi.org/10.1145/1367829.1367831 . .

J.S Vitter . . Random sampling with a reservoir . . ACMTrans. Math. Softw. , , 1985 . . 11 ( ( 1 ): ): 37 - - 57 . . DOI: 10.1145/3147.3165 http://doi.org/10.1145/3147.3165 . .

S Yoshida , , , T Morihara , , , H Yahagi , , , 等 . . Application of a word-based text compression method to Japanese and Chinese texts . . 1999 . . Data Compression Conf. . . 561 DOI: 10.1109/DCC.1999.785718 http://doi.org/10.1109/DCC.1999.785718 . .

J Ziv , , , A Lempel . . A universal algorithm for sequential data compression . . IEEE Trans. Inform. Theor. , , 1977 . . 23 ( ( 3 ): ): 337 - - 343 . . DOI: 10.1109/TIT.1977.1055714 http://doi.org/10.1109/TIT.1977.1055714 . .

浏览量

Downloads

CSCD

文章被引用时，请邮件提醒。

Submit

工具集

关联资源

SoftSSD: enabling rapid flash firmware prototyping for solid-state drives^&

Moving from exascale to zettascale computing: challenges and techniques

A novel non-volatile memory storage system for I/O-intensive applications

A reliable and energy-efficient storage system with erasure coding cache