SQLAnywhere
summary
SQL Anywhere is an embedded DBMS that runs from phones to Raspberry Pi to oil rigs, ships as a portable file and supports telepresence robot Ian Anywhere running for eight years; customers deploy over two million instances and databases from 20 MB to 500+ GB, with joins beyond 500 tables and up to 45,000 columns. It uses logical logging for redo/undo and point-in-time recovery, offers Mobile Link replication to backends, and applies adaptive optimizations including parametric plan caching and prefetching to trade optimizer cost against runtime unpredictability.
AI-written summary · about
chapters (21)
- 0:00Intro
- 2:37SAP SQL Anywhere Use Cases
- 9:15SQL Anywhere databases
- 11:19Changing a row in a table
- 12:40Query Processing Phases
- 15:17SAP SQL Anywhere Limits
- 17:15What makes SAP SQL Anywhere Installations and Product Unique?
- 22:09SQL Anywhere Key Features (3/3)
- 26:42SQL Anywhere Key Features - Synchronization
- 27:47Query Optimization Challenges (Anisoara Nica Inf. Syst 2012, ICE 2012/2011, SIGMOD 2009,...)
- 29:55Effects of optimizing at open time
- 32:16The pain of cheap requests
- 32:57Where is server time going?
- 37:26Access methods (Peter Bumbulis and Bowman, SIGMOD 2002)
- 41:43Load balancing multi-tenant workloads
- 44:00Automatic Statistics Management
- 45:25Column Histograms
- 46:19Motivation for Self Healing Statistics
- 46:45Dynamic Memory Management - Adjust Buffer Pool Size
- 47:24What's in the buffer pool?
- 51:24Adaptive Query Processing
related talks
48mSQLite – The everywhere database
34mOracle Autonomous Database: simplify your database management
45mForget SQL! Create your own AI Data Analyst
34mEdge & Offline-First App Data Management in 5G devices
24mgives databases to everyone
35mAmazon Aurora DSQL: Scalable, Distributed Relational Databases
30mUse SQLite in production
29mDatabase: What's new, what's next