The ACID BASE Conundrum
summary
Starts from CAP theorem and shows why distributed stores must trade consistency for availability, using a concrete example where x=5 gets writes x=10 and x=20 so locks in ACID stores block the second write while BASE stores let both nodes accept different values and converge to eventual consistency. Names implementations and mechanisms: Google Spanner and CockroachDB use distributed consensus like Paxos/Raft, run shared‑nothing clusters with in‑memory OLTP options, and let you tune consistency versus availability.
AI-written summary · about
chapters (7)
- 0:00Introduction
- 1:32What makes database choices hard
- 3:53ACID
- 5:38Base Compliance Stores
- 8:00NewSQL
- 9:17Summary
- 9:47Questions
related talks
52mTransactions: Myths, Surprises and Opportunities
01mDon't Drop ACID - Transactions in Distributed NoSQL
47mBeing transactional in 2019 - what does this actually mean?
48mThe Accord Consensus Protocol, how to do ACID transactions at scale
52mThe Computer Science behind a modern distributed data store
01h 00mDistributed transactions implementation trade-offs
58mDon’t Drop ACID - Transactions in Distributed NoSQL
01h 00mBeyond Eventual Consistency