kuguobing / awesome-distributed-systems

A curated list to learn about distributed systems

awesome-distributed-systems

A (hopefully) curated list on awesome material on distributed systems, inspired by other awesome frameworks like awesome-python. Most links will tend to be readings on architecture itself rather than code itself.

Bootcamp

Read things here before you start.

• CAP Theorem, Also plain english explanation
• Fallacies of Distributed Computing, expect things to break, everything
• Distributed systems theory for the distributed engineer, most of the papers/books in the blog might reappear in this list again. Still a good BFS approach to distributed systems.
• FLP Impossibility Result (paper), an easier blog post to follow along
• An Introduction to Distributed Systems @aphyr's excellent introduction to distributed systems

Books

• Distributed Systems for fun and profit [Free]
• Distributed Systems Principles and Paradigms, Andrew Tanenbaum [Amazon Link]
• Scalable Web Architecture and Distributed Systems [Free]
• Principles of Distributed Systems [Free] [ETH Zurich University]
• Making reliable distributed systems in the presence of software errors, [Free] Joe Amstrong's (Author of Erlang) PhD thesis
• Designing Data Intensive Applications [Amazon Link]
• Distributed Computing, By Hagit Attiya and Jennifer Welch
• Distributed Algorithms, Nancy Lynch [Amazon Link]
• Impossibility Results for Distributed Computing (paywall)

Papers

Must read papers on distributed systems. While nearly all of Lamport's work should feature here, just adding a few that must be read.

• Times, Clocks and Ordering of Events in Distributed Systems Lamport's paper, the Quintessential distributed systems primer
• Session Guarantees for Weakly Consistent Replicated Data a '94 paper that talks about various recommendations for session guarantees for eventually consistent systems, many of this would be standard vocabulary in reading other dist. sys papers, like monotonic reads, read your writes etc.

Storage & Databases

• Dynamo: Amazon's Highly Available Key Value Store Paraphrasing @fogus from their blog, it is very rare for a paper describing an active production system to influence the state of active research in any industry; this is one of those seminal distributed systems paper that solves the problem of a highly available and fault tolerant database in an elegant way, later paving the way for systems like Cassandra, and many other AP systems using a consistent hashing.
• Bigtable: A Distributed Storage System for Structured Data
• The Google File System
• Cassandra: A Decentralized Structured Storage System Inspired heavily by Dynamo, an now an open source
• CRUSH: Controlled, Scalable, Decentralized Placement of Replicated Data, the algorithm for the basis of Ceph distributed storage system, for the architecture itself read RADOS

Messaging systems

• The Log: What every software engineer should know about real-time data's unifying abstraction, a somewhat long read, but covers brilliantly on logs, which are at the heart of most distributed systems
• Kafka: a Distributed Messaging System for Log Processing

Distributed Consensus and Fault-Tolerance

• Practicle Byzantine Fault Tolerance
• The Byzantine Generals Problem
• Impossibility of Distributed Consensus with One Faulty Process
• The Part Time Parliament Paxos, Lamport's original Paxos paper, a bit difficult to understand, may require multiple passes
• Paxos Made Simple, a more terse readable Paxos paper by Lamport himself. Shorter and more easier compared to the original.
• The Chubby Lock Service for loosely coupled distributed systems Google's lock service used for loosely coupled distributed systems. Sort of Paxos as a Service for building other distributed systems. Primary inspiration behind other Service Discovery & Coordination tools like Zookeeper, etcd, Consul etc.
• Paxos made live - An engineering perspective Google's learning while implementing systems atop of Paxos. Demonstrates various practical issues encountered while implementing a theoritical concept.
• Raft Consensus Algorithm An alternative to Paxos for distributed consensus, that is much simpler to understand. Do checkout an interesting visualization of raft

Testing, monitoring and tracing

While designing distributed systems are hard enough, testing them is even harder.

• Dapper, Google's large scale distributed-systems tracing infrastructure, this was also the basis for the design of open source projects such as Zipkin, Pinpoint and HTrace.

Programming Models

• Distributed Programming Model
• PSync: a partially synchronous language for fault-tolerant distributed algorithms Video: Conference Video
• Programming Models for Distributed Computing
• Logic and Lattices for Distributed Programming

Verification of Distributed Systems

• Jepsen A framework for distributed systems verification, with fault injection @aphyr has featured enough times in this list already, but Jepsen and the blog posts that go with are a quintessntial addition to any distributed systems reading list.
• Verdi A Framework for Implementing and Formally Verifying Distributed Systems Paper

Courses

• Reliable Distributed Algorithms, Part 1, KTH Sweden
• Reliable Distributed Algorithms, Part 2, KTH Sweden
• Cloud Computing Concepts, University of Illinois
• CMU: Distributed Systems in Go Programming Language
• Software Defined Networking , Georgia Tech.
• ETH Zurich: Distributed Systems
• ETH Zurich: Distributed Systems Part 2, covers Distributed control algorithms, communication models, fault-tolerance among other things. In particular fault tolerence issues (models, consensus, agreement) and replication issues (2PC,3PC, Paxos), which are critical in understanding distributed systems are explained in great detail.

Blogs and other reading links

• Notes on Distributed Systems for Young Bloods
• High Scalability Several architectures of huge internet services, for eg twitter, whatsapp
• There is No Now, Problems with simultaneity in distributed systems
• Turing Lecture: The Computer Science of Concurrency: The Early Years, An article by Leslie Lamport on concurrency
• The Paper Trail blog, a very readable blog covering various aspects of distributed systems
• aphyr, Posts on jepsen series are pretty awesome
• All Things Distributed - Wernel Vogel's (Amazon CTO) blog on distributed systems
• Distributed Systems: Take Responsibility for Failover
• The C10K problem
• On Designing and Deploying Internet-Scale Services
• Files are hard A blog post on filesystem consistency, pretty important to read if you are into distributed storage or databases.
• Distributed Systems Testing: The Lost World Testing distributed systems are hard enough, a well researched blog post which again covers a lot of links to various approaches and other papers

Meta Lists

Other lists like this one

• Readings in distributed systems
• Distributed Systems meta list
• List of required readings for Distributed Systems Part of CMU's Engineering Distributed Systems course
• The Distributed Reader
• A Distributed Systems Reading List, A collection of material, mostly papers on Distributed Systems Theory as well as seminal industry papers
• Distributed Systems Readings, A comprehensive list of online courses related to distributed systems