Query Vectorization vs. Compilation in Advanced Database Systems - Lecture 10
Offered By: CMU Database Group via YouTube
Course Description
Overview
Explore the advanced database concepts of query vectorization and compilation in this lecture from Carnegie Mellon University's Advanced Database Systems course. Delve into Prof. Andy Pavlo's comprehensive analysis of these techniques, examining their implications, performance comparisons, and design considerations. Learn about single-threaded performance, evaluation methods, and the broader design space for database systems. Gain insights into ongoing projects, including DBdio, and understand their deliverables and presentation requirements. Discover the latest developments in database technology, from PostgreSQL to BPF, and how they relate to query optimization strategies.
Syllabus
Introduction
Summary
Vectorization
Query Compilation
Class Outline
Single System
Two Implications
Comparison
Single threaded performance
Performance counters
Evaluation
Design Space
Project 2 Overview
DBdio
Project 3 Overview
Project 3 deliverable
Project 3 presentation
Status update
Design document
Project showcase
Projects
The Germans
Hackers Delight
Postgres
BPF
Taught by
CMU Database Group
Related Courses
Scaling Postgres for Time Series Data with CitusMicrosoft via YouTube The Parquet Format and Performance Optimization Opportunities
Databricks via YouTube Powering Interactive Data Applications at Scale
Strange Loop Conference via YouTube Zedstore - Compressed Columnar Storage for Postgres
Linux Foundation via YouTube TiFlash - Make TiKV 10x Faster and HTAP-able
CNCF [Cloud Native Computing Foundation] via YouTube