分散式系統論文筆記目錄
04-18
理論
一致性模型
- Consistency in Non-Transactional Distributed Storage Systems
- Closing The Performance Gap between Causal Consistency and Eventual Consistency
- Weak Consistency: A Generalized Theory and Optimistic Implementations for Distributed Transactions
- Limitations of Highly-Available Eventually-Consistent Data Stores 筆記
- Linearizability: A Correctness Condition for Concurrent Objects
隔離性模型
- Highly Available Transactions: Virtues and Limitations
分散式時鐘
- Time, clocks, and the ordering of events in a distributed system 筆記
- Interval Tree Clocks
- Plausible clocks: Constant size logical clocks for distributed systems
- A Comparison Between Lamports Scalar Clocks and Singhal & Kshemkalyanis Vector Clocks
演算法
共識演算法
- Paxos Made Simple 筆記
- Revisiting the paxos algorithm
- Lower bounds for asynchronous consensus
- Fast Paxos
- Generalized Consensus and Paxos
- There Is More Consensus in Egalitarian Parliaments
以上是關於 Paxos 演算法的論文,此處應有個小結
- In Search of an Understandable Consensus Algorithm 筆記(含與 Paxos 的簡單對比)
- Planning for change in a formal verification of the raft consensus protocol
- Consensus: Bridging Theory And Practice
以上是關於 Raft 演算法的論文,此處應有個小結
- Impossibility of distributed consensus with one faulty process
- Brief Announcement: A Tight Space Bound for Consensus
- GentleRain : Cheap and Scalable Causal Consistency with Physical Clocks
- A comprehensive study of Convergent and Commutative Replicated Data Types 筆記
Erasure Coding
- Erasure Coding in Windows Azure Storage
分散式系統中間件
網路坐標系統
- Vivaldi: A Decentralized Network Coordinate System
- Network coordinates in the wild
失效檢測器
- The φ accrual failure detector 筆記
- Failure Detection in Large-Scale Systems: a Survey
成員管理
- SWIM: Scalable Weakly-consistent Infection-style process group Membership protocol 筆記
- Lifeguard : SWIM-ing with Situational Awareness
- Correctness of a gossip based membership protocol
- HyParView: A Membership Protocol for Reliable Gossip-Based Broadcast
分散式共享內存
- Sinfonia: a new paradigm for building scalable distributed systems 筆記
重疊網路
網路結構
- A survey of peer-to-peer content distribution technologies
- Peer to Peer Networks Management Survey
- The essence of P2P: A reference architecture for overlay networks
張帥:論文筆記:關於 P2P 的一些綜述
- Chord: A scalable content-addressable network
- A Scalable Content-Addressable Network
- Pastry: Scalable, Decentralized Object Location, and Routing for Large-Scale Peer-to-Peer Systems
- Tapestry: A global-scale overlay for rapid service deployment
- Kademlia: A Peer-to-Peer Information System Based on the XOR Metric
- DKS(N, k, f): a family of low communication, scalable and fault-tolerant infrastructures for P2P applications
此處應有一份關於以上論文的總結
- Skip graphs
網路通信
- Randomized rumor spreading
- Randomized rumor spreading revisited
- Brief Announcement: Optimal Address-Oblivious Epidemic Dissemination
張帥:論文筆記:[FOCS 2000] Randomized rumor spreading
- Scribe: A large-scale and decentralized application-level multicast infrastructure
- Application-Level Multicast using Content-Addressable Networks
- Ccag: Gossip Based Reliable Multicast Protocol
- Efficient and adaptive epidemic-style protocols for reliable and scalable multicast
分散式存儲
Key-Value 存儲
- Dynamo: Amazon』s Highly Available Key-value Store 筆記
- SILT: A Memory-Efficient, High-Performance Key-Value Store
- Efficient and Available In-memory KV-Store with Hybrid Erasure Coding and Replication
- Ambry: LinkedIns Scalable Geo-Distributed Object Store
- The RAMCloud Storage System
- Finding a needle in Haystack: facebooks photo storage 筆記
- F4: Facebooks Warm BLOB Storage System 筆記
- Windows Azure Storage : A Highly Available Cloud Storage Service with Strong Consistency
文件存儲
- The Google file system
- GFS: Evolution on Fast-Forward
- Colossus: Successor to the Google File System (GFS)
此處應有一份關於以上論文的總結
- The Hadoop Distributed File System
- Ceph: a scalable, high-performance distributed file system
- The quantcast file system
分散式資料庫
- Spanner: Googles Globally-Distributed Database
- Megastore: Providing Scalable, Highly Available Storage for Interactive Services
- F1: A Distributed SQL Database That Scales
- H-store: A High-Performance, Distributed Main Memory Transaction Processing System
時間序列資料庫
- Gorilla: A Fast, Scalable, In-Memory Time Series Database
- LittleTable: A Time-Series Database and Its Uses
列式存儲
結構化存儲
- Bigtable: A Distributed Storage System for Structured Data
- Cassandra: a decentralized structured storage system
分散式索引
- SLIK: scalable low-latency indexes for a key-value store
分散式計算
離線計算
MapReduce, Spark, etc.
在線計算
Streaming, OLAP, etc.
增量計算
- Large-scale Incremental Processing Using Distributed Transactions and Notifications
分散式緩存
memcached, tachyon, etc.
分散式資源管理
- Resource Management in Clouds: Survey and Research Challenges
- Apache Hadoop YARN: yet another resource negotiator
- Mesos: a platform for fine-grained resource sharing in the data center
- Omega: flexible, scalable schedulers for large compute clusters
- Large-scale cluster management at Google with Borg
- Firmament: Fast, Centralized Cluster Scheduling at Scale
- Apollo: Scalable and Coordinated Scheduling for Cloud-Scale Computing
分散式事務
- Consensus on Transaction Commit
- Omid , Reloaded : Scalable and Highly-Available Transaction Processing
推薦閱讀:
※分散式系統測試的應用方法——場景注入測試
※閱讀筆記:PowerGraph: Distributed Graph-Parallel Computation on Natural Graphs
※Elasticsearch分散式一致性原理剖析(一)-節點篇
※yaraft 的開發近況〔2017.11〕