Главная
Study mode:
on
1
Intro
2
Agenda
3
What is Vectorization
4
Why Vectorization Matters
5
Single Instruction Multiple Data
6
SIMD Example
7
Types of Vectorization
8
Types of Instructions
9
Streaming Instructions
10
Handling Exceptions
11
SIMD History
12
Tradeoffs
13
AVX 512
14
Wikipedia
15
Skylake 2017
16
Implementation
17
Automatic Vectorization
18
Automatic Vectorization Example
19
Driver Hints
20
C Restrictions
21
Explicit Vectorization
22
Memory Alignment
23
Permute
24
Out of Memory
25
Selective Store
26
Compress
27
Additional Values Span
28
Gather and Gather
29
Scatter
30
Vectorized Algorithms
31
MD Compare
32
Output Vector
33
Performance
34
Xeon Phi
35
Branchless
36
Vectorized
37
Memory Bandwidth
38
Invalid Tuples
39
Example
40
Materialization Model
41
Stage Buffer
42
Simple Pseudo Code
43
Prefetching
44
Results
45
Graph
46
Hash Tables
Description:
Explore a comprehensive lecture on vectorized query execution using SIMD in advanced database systems. Delve into the fundamentals of vectorization, its importance, and the concept of Single Instruction Multiple Data (SIMD). Examine various types of vectorization, SIMD instructions, and their historical development. Analyze implementation strategies, including automatic and explicit vectorization, memory alignment, and specialized instructions. Investigate vectorized algorithms, performance considerations, and the materialization model. Learn about stage buffers, prefetching, and hash tables in the context of vectorized query execution.

Vectorized Query Execution with SIMD - Lecture 8

CMU Database Group
Add to list
#Computer Science #High Performance Computing #Parallel Computing #SIMD #Computer Architecture #Instruction Set Architecture #CPU Architecture
0:00 / 0:00
1 of 46

Intro