20250219

Designing Data-Intensive Applications
- Detailed implementations or reasonings are a bit too difficult for me now. I prioritize going through the book for now, and I will go back to in-depth study later.
- Linerizability is a consistency model that ensures a system behaves as if there is only one copy of the data, with all operations applied atomicaly in a single, globally consistent order. While making distributed systems look as if a single-threaded program is appealing, its performance drawbacks make it an impractical option.
- Causal consistency offers a more practical alternative by ensuring that operations that are causally related are observed in the same order by all nodes, without enforcing a total global ordering. This allows for better performance and scalability while still maintaining meaningful consistency.
- However, capturing causal relationships using mechanisms like Lamport timestamps does not solve all coordination challenges, particularly in cases requiring strong agreement. This leads to the need for consensus algorithms to achieve a shared, reliable state across distributed nodes.
60 minutes walk
haircut
journaling (zero second thinking)
started a Go project
worked hard

TODO: