Главная
Study mode:
on
1
Intro
2
Marco Russo
3
What is VertiPaq?
4
Tabular query architecture
5
Row storage layout
6
Column storage layout
7
Querying a columnar database
8
Column vs row storage
9
Value encoding
10
Hash encoding
11
Run Length Encoding (RLE)
12
Compression of one column
13
VertiPaq compression
14
Reducing column size
15
Segmentation
16
Processing phases
17
Special Case of 3rd Segment
18
Calculated columns and tables
19
Query parallelization
20
Hierarchies
21
Memory Usage During Process
22
Data memory usage
23
VertiPaq Analyzer
24
Query memory usage
25
Early materialization
26
Late materialization
Description:
Dive deep into the VertiPaq engine's architecture and functionality in this comprehensive 1-hour 10-minute conference talk from PASS Data Community Summit. Explore the columnar database used by SQL Server Analysis Services Tabular, Power BI, and Power Pivot, known for its impressive performance in speed and compression. Discover how VertiPaq stores information, gaining valuable insights into its inner workings and learning optimal data warehouse modeling techniques. Examine common and effective methods to enhance compression ratios and boost performance in Tabular data models. Cover topics such as row and column storage layouts, querying techniques, value encoding, hash encoding, Run Length Encoding (RLE), segmentation, processing phases, calculated columns and tables, query parallelization, hierarchies, memory usage, and materialization strategies.

Inside the VertiPaq Engine

PASS Data Community Summit
Add to list