Introduction
It is my favorite book chapter. I have good time to learn something from the book chapter. I just start to read the book chapter less than one week ago, I tried a few times, each time I had to go back to previous chapters, and then learned some basics first. I like to take some notes for each subtitle in the book chapter.
Data Integration
Unbunding Databases
Aiming for Correctness
Doing the right thing
Storage engine - log-structured storage, B-Trees, and column-oriented storage - Chapter 3replication - single leader, multi-leader and leaderless approaches - Chapter 5
Durable system record - ?
Conversely, search indexes are generally not very suitable as a durable system of record, and so many applications need to combine two different tools in order to satisfy all of the requirements.
Page 492
At an abstract level, they achieve a similar goal by different means. Distributed transactions
decide on an ordering of writes by using locks for mutual exclusion (see
“Two-Phase Locking (2PL)” on page 257), while CDC and event sourcing use a log
for ordering. Distributed transactions use atomic commit to ensure that changes take
effect exactly once, while log-based systems are often based on deterministic retry
and idempotence.
Abstract level
Distributed transactions
ordering of writes by using locks from mutual exclusion
CDC and event sourcing
use a log for ordering
Statement:
Distributed transactions use atomic commit to ensure that changes take effect exactly once, while log-based systems are often based on deterministic retry and idempotence.
In “Aiming for Correctness” on page 515 we will discuss some approaches for implementing
stronger guarantees on top of asynchronously derived systems, and work
toward a middle ground between distributed transactions and asynchronous logbased
systems.
No comments:
Post a Comment