20250111

• 0秒思考
• Designing Data-Intensive Applications
  • write skew: a race condition where multiple transactions read the same objects and update some of them concurrently, leading to an inconsistent state. In constrast, a dirty write or lost update occurs when two transactions simultaneously read and update the same object, resulting in one update being overwritten and lost.
    • For example, you can think of a meeting room booking system. When two transactions that read bookings to ensure the the room is empty at a specific time and add a new booking record occur almost at the same time, both transactions may add their bookings that conflict. In this case, simply getting a lock for a record to be updated is infeasible.
      • “This effect, where a write in one transaction changes the result of a search query in another transaction, is called phantom”.
    • One way to address the issue is materializing conflicts, which turns a phantom into a lock conflict by creating concrete set of rows that can be dealt with a lock.
      • For example, we can create records (mtg room, time period) beforehand, and use a lock when updating the records.
      • This is uncommon in reality because it’s difficult to think about how to materialize conflicts.
  • serializable isolation: considered to be the strongest isolation level
    • “It guarantees that even though transactions may execute in parallel, the end result is the same as if they had executed one at a time, serially, without any concurrency.”
• I provided a consultation for a couple of my friend’s couple.
• Civics Questions
  • Mike Johnson

2 hours walk (exploration)

TODO:

• 2044E, F
• 2048D
• Smog test
  • Ask my parents to transfer the ownership of my car
• Naturalization Interview (20250207)

index 20250110 20250112