Scaling problems • Partitioning or sharing works to an extent
9
Avoid sharing mutable data
10
questions about data access
11
Non-primary key access - Finding items by value, not by key
12
Range scans and sequential access - Requires ordering Leoperators, ordering costs • Requires iberators/cursors/traversal state
13
Read/write ratio
14
Working set size and skew • How much of the common data will fit in main
15
Consistency • Do all copies of the data need to be exactly up- to-date right now
16
Reader/writer vs data flow
17
Sync vs async systems
18
Data flow and sync/async
19
Content management example
20
Larger example
21
Command Query Representation Separation (CORS)
22
CORS examples
Description:
Explore the differences between reading and writing data across various topics in software development in this ACCU Conference talk. Delve into high-level architecture, low-level concurrency, REST APIs, functional programming, security, performance, and correctness. Discover how separating these concepts can lead to more optimal system designs and broader perspectives. Learn about scaling challenges, data partitioning, avoiding shared mutable data, and handling non-primary key access. Examine read/write ratios, working set sizes, consistency issues, and the distinctions between reader/writer and data flow models. Investigate synchronous vs asynchronous systems, content management examples, and the Command Query Representation Separation (CQRS) pattern. Gain insights into alternative approaches that offer more fruitful design considerations for building robust and efficient software systems.