Indexing 7: v-byte encoding (compression)

preview_player
Показать описание
V-byte encoding allows us to use fewer bits to represent smaller numbers (which result from delta encoding), while allowing arbitrarily large numbers in the index. We use 7 bits of each byte to represent the number itself, and 1 bit to indicate how many bytes are used to represent the full number.
  1. Victor Lavrenko
  2. tts
  3. text
  4. text technologies
  5. retrieval
  6. information retrieval
Рекомендации по теме
Indexing 7: v-byte 0:09:09

Indexing 7: v-byte encoding (compression)

14.249 7Bit Encoding 0:09:44

14.249 7Bit Encoding

Indexing 6: delta 0:05:58

Indexing 6: delta encoding (compression)

08 07 Index 0:06:30

08 07 Index Compression

indexing 7 0:33:37

indexing 7

COMP526 7-5 §7.4 0:26:37

COMP526 7-5 §7.4 Run length encoding

DW M7 RL 0:29:10

DW M7 RL 7 7 2 Compression of Bitmap Indexes

Index Compression in 0:05:03

Index Compression in Oracle Database 12c Release 1 (12.1.0.2)

7.1. Index compression 0:16:47

7.1. Index compression - preliminaries

7.3. Index compression 0:11:59

7.3. Index compression - dictionary compression

1.2.7 Variable-length Encoding 0:03:40

1.2.7 Variable-length Encoding

IR Course Lecture 0:33:43

IR Course Lecture 9: Index Compression

Engineering Fast Indexes 0:30:10

Engineering Fast Indexes for Big Data Applications: Spark Summit East talk by Daniel Lemire

Advanced Data Structures: 0:04:25

Advanced Data Structures: Data Compression

Compression Algorithm Demo 0:00:34

Compression Algorithm Demo

COMP526 Unit 7-7 0:20:05

COMP526 Unit 7-7 2020-03-24 Compression: Burrows-Wheeler-Transform

Byte Alignment Implementation 0:09:03

Byte Alignment Implementation : Bit Manipulation Interview Question.

Compression 0:07:52

Compression

8. Unary coding 0:01:30

8. Unary coding

COMP526 Unit 7-5b 0:30:03

COMP526 Unit 7-5b 2020-03-24 Compression: Lempel-Ziv-Welche decoding

Variable Length Codes 0:20:16

Variable Length Codes (Ep 1, Compressor Head)

Compression using Run 0:03:33

Compression using Run Length Encoding

Class 7: Big 1:26:38

Class 7: Big Data (CC5212) - Information Retrieval: Crawling & Indexing

09 - Database 1:20:24

09 - Database Compression (CMU Databases / Spring 2020)

Комментарии
Автор

Very useful! Came here after reading the chapter of Information Retrieval book, was a little bit confused. Glad you cleared it up!

ihorprotsenko
Автор

at 3:08, you said a byte start with 0 means that this is the first byte of a multi-byte number. I don't think it is accurate. It should be: a byte start with 0 means that this is NOT the last byte of a multi-byte number and we should continue looking next byte.
Please correct me if I am wrong. Thanks!

zjuzhanxf
Автор

Is this also how variable length encoding like utf-8 works? 🤔

hansisbrucker
Автор

Thanks this was awesome explanation 😁😀

RAJATTHEPAGAL