有同事找我推荐阅读文章，整理了一下，也放在博客上吧

单机存储引擎：
目前工程上比较靠谱的无锁算法：Hazard Pointers: Safe Memory Reclamation for Lock-Free Objects
B树与读写并发：B-trees, Shadowing, and Clones
单机引擎：The Bw-Tree: A B-tree for New Hardware Platforms
单机引擎：The Log-Structured Merge-Tree (LSM-Tree)
LSM变种：WiscKey: Separating Keys from Values in SSD-conscious Storage
前缀压测引擎（terark重要参考）：SuRF: Practical Range Query Filtering with Fast Succinct Tries


分布式存储：
文件系统：The Google File System，第二代为colossus没有论文
对象存储：Finding a needle in Haystack: Facebook’s photo storage

分布式数据库：
最有名的bigtable：Bigtable: A Distributed Storage System for Structured Data
分布式事务系统（tidb重要参考）：Large-scale Incremental Processing Using Distributed Transactions and Notifications
分布式数据库：Spanner: Google’s Globally-Distributed Database
一致性hash的kv存储：Dynamo: Amazon’s Highly Available Key-value Store
分布式数据库：Amazon Aurora: Design Considerations for High Throughput Cloud-Native Relational Databases
伯克利最新KV：Anna: A KVS For Any Scale

一致性协议：
Paxos原始论文：The Part-Time Parliament
原始论文读不懂可以先看这篇：Paxos Made Simple
Raft论文：In Search of an Understandable Consensus Algorithm，导演未删节版：CONSENSUS: BRIDGING THEORY AND PRACTICE
分布式事务：Consensus on Transaction Commit
集中排序的分布式事务处理方式（虽然我不是很喜欢）：Calvin: Fast Distributed Transactions for Partitioned Database Systems
混合逻辑时钟（用于分布式事务排序，cockroachdb重要参考）：Logical Physical Clocks and Consistent Snapshots in Globally Distributed Databases
SQL隔离级别：A Critique of ANSI SQL Isolation Levels

Paxos也可以看看这里的Paxos三部曲