Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

August 2013 HUG: Compression Options in Hadoop - A Tale of Tradeoffs


Published on

Yahoo! is one of the most-visited web sites in the world. It runs one of the largest private cloud infrastructures, one that operates on petabytes of data every day. Being able to store and manage that data well is essential to the efficient functioning of Yahoo!`s Hadoop clusters. A key component that enables this efficient operation is data compression. With regard to compression algorithms, there is an underlying tension between compression ratio and compression performance. Consequently, Hadoop provides support for several compression algorithms, including gzip, bzip2, Snappy, LZ4 and others. This plethora of options can make it difficult for users to select appropriate codecs for their MapReduce jobs. This talk attempts to provide guidance in that regard. Performance results with Gridmix and with several corpuses of data are presented. The talk also describes enhancements we have made to the bzip2 codec that improve its performance. This will be of particular interest to the increasing number of users operating on “Big Data” who require the best possible ratios. The impact of using the Intel IPP libraries is also investigated; these have the potential to improve performance significantly. Finally, a few proposals for future enhancements to Hadoop in this area are outlined.

Speaker: Govind Kamat, Member of Technical Staff, Yahoo!

Published in: Technology
  • Be the first to comment

August 2013 HUG: Compression Options in Hadoop - A Tale of Tradeoffs

  1. 1. Compression Options In Hadoop – A Tale of Tradeoffs Govind Kamat 39th Bay Area Hadoop Users Group (HUG) Meetup Yahoo! URL’s Café Sunnyvale, CA August 21, 2013
  2. 2. Introduction 2 Sumeet Singh Director of Products, Hadoop Cloud Engineering Group 701 First Avenue Sunnyvale, CA 94089 USA Govind Kamat Technical Yahoo!, Hadoop Cloud Engineering Group §  Member of Technical Staff in the Hadoop Services team at Yahoo! §  Focuses on HBase and Hadoop performance §  Worked with the Performance Engineering Group on improving the performance and scalability of several Yahoo! applications §  Experience includes development of large-scale software systems, microprocessor architecture, instruction-set simulators, compiler technology and electronic design 701 First Avenue Sunnyvale, CA 94089 USA §  Leads Hadoop products team at Yahoo! §  Responsible for Product Management, Customer Engagements, Evangelism, and Program Management §  Prior to this role, led Strategy functions for the Cloud Platform Group at Yahoo!
  3. 3. Agenda 3 Data Compression in Hadoop1 Available Compression Options2 Understanding and Working with Compression Options3 Problems Faced at Yahoo! with Large Data Sets4 Performance Evaluations, Native Bzip2, and IPP Libraries5 Wrap-up and Future Work6
  4. 4. Compression Needs and Tradeoffs in Hadoop 4 §  Storage §  Disk I/O §  Network bandwidth §  CPU Time §  Hadoop jobs are data-intensive, compressing data can speed up the I/O operations §  MapReduce jobs are almost always I/O bound §  Compressed data can save storage space and speed up data transfers across the network §  Capital allocation for hardware can go further §  Reduced I/O and network load can bring significant performance improvements §  MapReduce jobs can finish faster overall §  On the other hand, CPU utilization and processing time increases during compression and decompression §  Understanding the tradeoffs is important for MapReduce pipeline’s overall performance The Compression Tradeoff
  5. 5. Data Compression in Hadoop’s MR Pipeline 5 Input splits Map Source: Hadoop: The Definitive Guide, Tom White Output ReduceBuffer in memory Partition and Sort fetch Merge on disk Merge and sort Other maps Other reducers I/P compressed Mapper decompresses Mapper O/P compressed 1 Map Reduce Reduce I/P Map O/P Reducer I/P decompresses Reducer O/P compressed 2 3 Sort & Shuffle Compress Decompress
  6. 6. Compression Options in Hadoop (1/2) 6 Format Algorithm Strategy Emphasis Comments zlib Uses DEFLATE (LZ77 and Huffman coding) Dictionary-based, API Compression ratio Default codec gzip Wrapper around zlib Dictionary-based, standard compression utility Same as zlib, codec operates on and produces standard gzip files For data interchange on and off Hadoop bzip2 Burrows-Wheeler transform Transform-based, block-oriented Higher compression ratios than zlib Common for Pig LZO Variant of LZ77 Dictionary-based, block-oriented, API High compression speeds Common for intermediate compression, HBase tables LZ4 Simplified variant of LZ77 Fast scan, API Very high compression speeds Available in newer Hadoop distributions Snappy LZ77 Block-oriented, API Very high compression speeds Came out of Google, previously known as Zippy
  7. 7. Compression Options in Hadoop (2/2) 7 Format Codec (Defined in io.compression.codecs) File Extn. Splittable Java/ Native zlib/ DEFLATE (default) !.deflate! N Y/ Y gzip ! .gz! N Y/ Y bzip2 ! .bz2! Y Y/ Y LZO (download separately) com.hadoop.compression.lzo.LzoCodec ! .lzo! N N/ Y LZ4 ! .lz4! N N/ Y Snappy ! .snappy! N N/ Y NOTES: §  Splittability – Bzip2 is “splittable”, can be decompressed in parallel by multiple MapReduce tasks. Other algorithms require all blocks together for decompression with a single MapReduce task. §  LZO – Removed from Hadoop because the LZO libraries are licensed under the GNU GPL. LZO format is still supported and the codec can be downloaded separately and enabled manually. §  Native bzip2 codec – added by Yahoo! as part of this work in Hadoop 0.23
  8. 8. Space-Time Tradeoff of Compression Options 8 64%, 32.3 71%, 60.0 47%, 4.842%, 4.0 44%, 2.4 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 40% 45% 50% 55% 60% 65% 70% 75% CPUTimeinSec. (Compress+Decompress) Space Savings Bzip2 Zlib (Deflate, Gzip) LZOSnappy LZ4 Note: A 265 MB corpus from Wikipedia was used for the performance comparisons. Space savings is defined as [1 – (Compressed/ Uncompressed)] Codec Performance on the Wikipedia Text Corpus High Compression Ratio High Compression Speed
  9. 9. Using Data Compression in Hadoop 9 Phase in MR Pipeline Config Values Input data to Map File extension recognized automatically for decompression File extensions for supported formats Note: For SequenceFile, headers have the information [compression (boolean), block compression (boolean), and compression codec] One of the supported codecs one defined in io.compression.codecs! Intermediate (Map) Output! false (default), true! ! one defined in io.compression.codecs! Final (Reduce) Output mapreduce.output.fileoutputformat. compress! false (default), true mapreduce.output.fileoutputformat. compress.codec! one defined in io.compression.codecs! mapreduce.output.fileoutputformat. compress.type! Type of compression to use for SequenceFile outputs: NONE, RECORD (default), BLOCK 1 2 3
  10. 10. §  Compress the input data, if large §  Always use compression, particularly if spillage or slow network transfers §  Compress for storage/ archival, better write speeds, or between MR jobs §  Use splittable algo such as bzip2, or use zlib with SequenceFile format §  Use faster codecs such as LZO, LZ4, or Snappy §  Use standard utility such as gzip or bzip2 for data interchange, and faster codecs for chained jobs When to Use Compression and Which Codec 10 Map ReduceShuffle & Sort Input data to Map Intermediate (Map) Output I/P compressed Mapper decompresses Mapper O/P compressed 1 Reducer I/P decompresses Reducer O/P compressed 2 3 Compress Decompress Final Reduce Output
  11. 11. Compression in the Hadoop Ecosystem 11 Component When to Use What to Use Pig §  Compressing data between MR job §  Typical in Pig scripts that include joins or other operators that expand your data size Enable compression and select the codec: pig.tmpfilecompression = true! pig.tmpfilecompression.codec = gzip, lzo! ! Hive §  Intermediate files produced by Hive between multiple map- reduce jobs §  Hive writes output to a table Enable intermediate or output compression: hive.exec.compress.intermediate = true! hive.exec.compress.output = true! HBase §  Compress data at the CF level (support for LZO, gzip, Snappy, and LZ4) List required JNI libraries: hbase.regionserver.codecs! ! Enabling compression: create ’table', { NAME => 'colfam', COMPRESSION => ’LZO' }! alter ’table', { NAME => 'colfam', COMPRESSION => ’LZO' } !
  12. 12. 4.2M Jobs, Jun 10-16, 2013 Compression in Hadoop at Yahoo! 12 99.8% 0.2% LZO 98.3% gzip 1.1% zlib / default 0.5% bzip2 0.1% Map ReduceShuffle & Sort Input data to Map Intermediate (Map) Output 1 2 3 Final Reduce Output 39.0% 61.0% LZO 55% gzip 35% bzip2 5% zlib / default 5% 4.2M Jobs, Jun 10-16, 2013 98% 2% zlib / default 73% gzip 22% bzip2 4% LZO 1% 380M Files on Jun 16, 2013 (/data, /projects) Includes intermediate Pig/ Hive compression Pig Intermediate Compressed
  13. 13. Compression for Data Storage Efficiency §  DSE considerations at Yahoo! §  RCFile instead of SequenceFile §  Faster implementation of bzip2 §  Native-code bzip2 codec §  HADOOP-84621, available in 0.23.7 §  Substituting the IPP library 13 1 Native-code bzip2 implementation done in collaboration with Jason Lowe, Hadoop Core PMC member
  14. 14. IPP Libraries §  Integrated Performance Primitives from Intel §  Algorithmic and architectural optimizations §  Processor-specific variants of each function §  Applications remain processor-neutral §  Compression: LZ, RLE, BWT, LZO §  High level formats include: zlib, gzip, bzip2 and LZO 14
  15. 15. Measuring Standalone Performance §  Standard programs (gzip, bzip2) used §  Driver program written for other cases §  32-bit mode §  Single-threaded §  JVM load overhead discounted §  Default compression level §  Quad-core Xeon machine 15
  16. 16. Data Corpuses Used §  Binary files §  Generated text from randomtextwriter §  Wikipedia corpus §  Silesia corpus 16
  17. 17. Compression Ratio 0 50 100 150 200 250 300 uncomp zlib bzip2 LZO Snappy LZ4 FileSize(MB) exe rtext wiki silesia 17
  18. 18. Compression Performance 29 23 63 44 26 0 10 20 30 40 50 60 70 80 90 zlib IPP-zlib Java-bzip2 bzip2 IPP-bzip2 CPUTime(sec) exe rtext wiki silesia 18
  19. 19. Compression Performance (Fast Algorithms) 3.2 2.9 1.7 0 0.5 1 1.5 2 2.5 3 3.5 LZO Snappy LZ4 CPUTime(sec) exe rtext wiki silesia 19
  20. 20. Decompression Performance 3 2 21 17 12 0 5 10 15 20 25 zlib IPP-zlib Java-bzip2 bzip2 IPP-bzip2 CPUTime(sec) exe rtext wiki silesia 20
  21. 21. Decompression Performance (Fast Algorithms) 1.6 1.1 0.7 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 LZO Snappy LZ4 CPUTime(sec) exe rtext wiki silesia 21
  22. 22. Compression Performance within Hadoop §  Daytona performance framework §  GridMix v1 §  Loadgen and sort jobs §  Input data compressed with zlib / bzip2 §  LZO used for intermediate compression §  35 datanodes, dual-quad-core machines 22
  23. 23. Map Performance 47 46 46 33 0 5 10 15 20 25 30 35 40 45 50 Java-bzip2 bzip2 IPP-bzip2 zlib MapTime(sec) 23
  24. 24. Reduce Performance 31 28 18 14 0 5 10 15 20 25 30 35 Java-bzip2 bzip2 IPP-bzip2 zlib ReduceTime(min) 24
  25. 25. Job Performance 38 34 23 19 38 34 25 18 0 5 10 15 20 25 30 35 40 Java-bzip2 bzip2 IPP-bzip2 zlib JobTime(min) sort loadgen 25
  26. 26. Future Work §  Splittability support for native-code bzip2 codec §  Enhancing Pig to use common bzip2 codec §  Optimizing the JNI interface and buffer copies §  Varying the compression effort parameter §  Performance evaluation for 64-bit mode §  Updating the zlib codec to specify alternative libraries §  Other codec combinations, such as zlib for transient data §  Other compression algorithms 26
  27. 27. Considerations in Selecting Compression Type §  Nature of the data set §  Chained jobs §  Data-storage efficiency requirements §  Frequency of compression vs. decompression §  Requirement for compatibility with a standard data format §  Splittability requirements §  Size of the intermediate and final data §  Alternative implementations of compression libraries 27