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.
Roaring Bitmaps (January 2016)
Roaring Bitmaps (January 2016)
Roaring Bitmaps (January 2016)
Roaring Bitmaps (January 2016)
Roaring Bitmaps (January 2016)
Roaring Bitmaps (January 2016)
Roaring Bitmaps (January 2016)
Roaring Bitmaps (January 2016)
Roaring Bitmaps (January 2016)
Roaring Bitmaps (January 2016)
Roaring Bitmaps (January 2016)
Roaring Bitmaps (January 2016)
Roaring Bitmaps (January 2016)
Roaring Bitmaps (January 2016)
Roaring Bitmaps (January 2016)
Roaring Bitmaps (January 2016)
Roaring Bitmaps (January 2016)
Roaring Bitmaps (January 2016)
Roaring Bitmaps (January 2016)
Roaring Bitmaps (January 2016)
Roaring Bitmaps (January 2016)
Roaring Bitmaps (January 2016)
Roaring Bitmaps (January 2016)
Roaring Bitmaps (January 2016)
Roaring Bitmaps (January 2016)
Roaring Bitmaps (January 2016)
Roaring Bitmaps (January 2016)
Roaring Bitmaps (January 2016)
Roaring Bitmaps (January 2016)
Roaring Bitmaps (January 2016)
Roaring Bitmaps (January 2016)
Roaring Bitmaps (January 2016)
Roaring Bitmaps (January 2016)
Roaring Bitmaps (January 2016)
Upcoming SlideShare
Loading in …5
×

of

Roaring Bitmaps (January 2016) Slide 1 Roaring Bitmaps (January 2016) Slide 2 Roaring Bitmaps (January 2016) Slide 3 Roaring Bitmaps (January 2016) Slide 4 Roaring Bitmaps (January 2016) Slide 5 Roaring Bitmaps (January 2016) Slide 6 Roaring Bitmaps (January 2016) Slide 7 Roaring Bitmaps (January 2016) Slide 8 Roaring Bitmaps (January 2016) Slide 9 Roaring Bitmaps (January 2016) Slide 10 Roaring Bitmaps (January 2016) Slide 11 Roaring Bitmaps (January 2016) Slide 12 Roaring Bitmaps (January 2016) Slide 13 Roaring Bitmaps (January 2016) Slide 14 Roaring Bitmaps (January 2016) Slide 15 Roaring Bitmaps (January 2016) Slide 16 Roaring Bitmaps (January 2016) Slide 17 Roaring Bitmaps (January 2016) Slide 18 Roaring Bitmaps (January 2016) Slide 19 Roaring Bitmaps (January 2016) Slide 20 Roaring Bitmaps (January 2016) Slide 21 Roaring Bitmaps (January 2016) Slide 22 Roaring Bitmaps (January 2016) Slide 23 Roaring Bitmaps (January 2016) Slide 24 Roaring Bitmaps (January 2016) Slide 25 Roaring Bitmaps (January 2016) Slide 26 Roaring Bitmaps (January 2016) Slide 27 Roaring Bitmaps (January 2016) Slide 28 Roaring Bitmaps (January 2016) Slide 29 Roaring Bitmaps (January 2016) Slide 30 Roaring Bitmaps (January 2016) Slide 31 Roaring Bitmaps (January 2016) Slide 32 Roaring Bitmaps (January 2016) Slide 33 Roaring Bitmaps (January 2016) Slide 34
Upcoming SlideShare
Compressing column-oriented indexes
Next
Download to read offline and view in fullscreen.

5 Likes

Share

Download to read offline

Roaring Bitmaps (January 2016)

Download to read offline

Better bitmap performance with Roaring bitmaps

Bitmaps are used to implement fast set operations in software. They are frequently found in databases and search engines.

Without compression, bitmaps scale poorly, so they are often compressed. Many bitmap compression techniques have been proposed, almost all relying primarily on run-length encoding (RLE). For example, Oracle relies on BBC bitmap compression while the version control system Git and Apache Hive rely on EWAH compression.

We can get superior performance with a hybrid compression technique that uses both uncompressed bitmaps and packed arrays inside a two-level tree. An instance of this technique, Roaring, has been adopted by several production platforms (e.g., Apache Lucene/Solr/Elastic, Apache Spark, eBay's Apache Kylin and Metamarkets' Druid).

Overall, our implementation of Roaring can be several times faster (up to two orders of magnitude) than the implementations of traditional RLE-based alternatives (WAH, Concise, EWAH) while compressing better. We review the design choices and optimizations that make these good results possible.

Related Books

Free with a 30 day trial from Scribd

See all
  • coalash

    Jan. 18, 2019
  • ssuser0b4a67

    Jan. 12, 2019
  • KhoaNguyen136

    Sep. 17, 2016
  • jimmyhit

    May. 15, 2016
  • vvhung

    Apr. 7, 2016

Better bitmap performance with Roaring bitmaps Bitmaps are used to implement fast set operations in software. They are frequently found in databases and search engines. Without compression, bitmaps scale poorly, so they are often compressed. Many bitmap compression techniques have been proposed, almost all relying primarily on run-length encoding (RLE). For example, Oracle relies on BBC bitmap compression while the version control system Git and Apache Hive rely on EWAH compression. We can get superior performance with a hybrid compression technique that uses both uncompressed bitmaps and packed arrays inside a two-level tree. An instance of this technique, Roaring, has been adopted by several production platforms (e.g., Apache Lucene/Solr/Elastic, Apache Spark, eBay's Apache Kylin and Metamarkets' Druid). Overall, our implementation of Roaring can be several times faster (up to two orders of magnitude) than the implementations of traditional RLE-based alternatives (WAH, Concise, EWAH) while compressing better. We review the design choices and optimizations that make these good results possible.

Views

Total views

3,142

On Slideshare

0

From embeds

0

Number of embeds

49

Actions

Downloads

58

Shares

0

Comments

0

Likes

5

×